-
ZePo: Zero-Shot Portrait Stylization with Faster Sampling
Authors:
Jin Liu,
Huaibo Huang,
Jie Cao,
Ran He
Abstract:
Diffusion-based text-to-image generation models have significantly advanced the field of art content synthesis. However, current portrait stylization methods generally require either model fine-tuning based on examples or the employment of DDIM Inversion to revert images to noise space, both of which substantially decelerate the image generation process. To overcome these limitations, this paper p…
▽ More
Diffusion-based text-to-image generation models have significantly advanced the field of art content synthesis. However, current portrait stylization methods generally require either model fine-tuning based on examples or the employment of DDIM Inversion to revert images to noise space, both of which substantially decelerate the image generation process. To overcome these limitations, this paper presents an inversion-free portrait stylization framework based on diffusion models that accomplishes content and style feature fusion in merely four sampling steps. We observed that Latent Consistency Models employing consistency distillation can effectively extract representative Consistency Features from noisy images. To blend the Consistency Features extracted from both content and style images, we introduce a Style Enhancement Attention Control technique that meticulously merges content and style features within the attention space of the target image. Moreover, we propose a feature merging strategy to amalgamate redundant features in Consistency Features, thereby reducing the computational load of attention control. Extensive experiments have validated the effectiveness of our proposed framework in enhancing stylization efficiency and fidelity. The code is available at \url{https://github.com/liujin112/ZePo}.
△ Less
Submitted 10 August, 2024;
originally announced August 2024.
-
VITA: Towards Open-Source Interactive Omni Multimodal LLM
Authors:
Chaoyou Fu,
Haojia Lin,
Zuwei Long,
Yunhang Shen,
Meng Zhao,
Yifan Zhang,
Xiong Wang,
Di Yin,
Long Ma,
Xiawu Zheng,
Ran He,
Rongrong Ji,
Yunsheng Wu,
Caifeng Shan,
Xing Sun
Abstract:
The remarkable multimodal capabilities and interactive experience of GPT-4o underscore their necessity in practical applications, yet open-source models rarely excel in both areas. In this paper, we introduce VITA, the first-ever open-source Multimodal Large Language Model (MLLM) adept at simultaneous processing and analysis of Video, Image, Text, and Audio modalities, and meanwhile has an advance…
▽ More
The remarkable multimodal capabilities and interactive experience of GPT-4o underscore their necessity in practical applications, yet open-source models rarely excel in both areas. In this paper, we introduce VITA, the first-ever open-source Multimodal Large Language Model (MLLM) adept at simultaneous processing and analysis of Video, Image, Text, and Audio modalities, and meanwhile has an advanced multimodal interactive experience. Starting from Mixtral 8x7B as a language foundation, we expand its Chinese vocabulary followed by bilingual instruction tuning. We further endow the language model with visual and audio capabilities through two-stage multi-task learning of multimodal alignment and instruction tuning. VITA demonstrates robust foundational capabilities of multilingual, vision, and audio understanding, as evidenced by its strong performance across a range of both unimodal and multimodal benchmarks. Beyond foundational capabilities, we have made considerable progress in enhancing the natural multimodal human-computer interaction experience. To the best of our knowledge, we are the first to exploit non-awakening interaction and audio interrupt in MLLM. VITA is the first step for the open-source community to explore the seamless integration of multimodal understanding and interaction. While there is still lots of work to be done on VITA to get close to close-source counterparts, we hope that its role as a pioneer can serve as a cornerstone for subsequent research. Project Page: https://vita-home.github.io.
△ Less
Submitted 9 August, 2024;
originally announced August 2024.
-
Fairness and Bias Mitigation in Computer Vision: A Survey
Authors:
Sepehr Dehdashtian,
Ruozhen He,
Yi Li,
Guha Balakrishnan,
Nuno Vasconcelos,
Vicente Ordonez,
Vishnu Naresh Boddeti
Abstract:
Computer vision systems have witnessed rapid progress over the past two decades due to multiple advances in the field. As these systems are increasingly being deployed in high-stakes real-world applications, there is a dire need to ensure that they do not propagate or amplify any discriminatory tendencies in historical or human-curated data or inadvertently learn biases from spurious correlations.…
▽ More
Computer vision systems have witnessed rapid progress over the past two decades due to multiple advances in the field. As these systems are increasingly being deployed in high-stakes real-world applications, there is a dire need to ensure that they do not propagate or amplify any discriminatory tendencies in historical or human-curated data or inadvertently learn biases from spurious correlations. This paper presents a comprehensive survey on fairness that summarizes and sheds light on ongoing trends and successes in the context of computer vision. The topics we discuss include 1) The origin and technical definitions of fairness drawn from the wider fair machine learning literature and adjacent disciplines. 2) Work that sought to discover and analyze biases in computer vision systems. 3) A summary of methods proposed to mitigate bias in computer vision systems in recent years. 4) A comprehensive summary of resources and datasets produced by researchers to measure, analyze, and mitigate bias and enhance fairness. 5) Discussion of the field's success, continuing trends in the context of multimodal foundation and generative models, and gaps that still need to be addressed. The presented characterization should help researchers understand the importance of identifying and mitigating bias in computer vision and the state of the field and identify potential directions for future research.
△ Less
Submitted 5 August, 2024;
originally announced August 2024.
-
Individualized multi-horizon MRI trajectory prediction for Alzheimer's Disease
Authors:
Rosemary He,
Gabriella Ang,
Daniel Tward
Abstract:
Neurodegeneration as measured through magnetic resonance imaging (MRI) is recognized as a potential biomarker for diagnosing Alzheimer's disease (AD), but is generally considered less specific than amyloid or tau based biomarkers. Due to a large amount of variability in brain anatomy between different individuals, we hypothesize that leveraging MRI time series can help improve specificity, by trea…
▽ More
Neurodegeneration as measured through magnetic resonance imaging (MRI) is recognized as a potential biomarker for diagnosing Alzheimer's disease (AD), but is generally considered less specific than amyloid or tau based biomarkers. Due to a large amount of variability in brain anatomy between different individuals, we hypothesize that leveraging MRI time series can help improve specificity, by treating each patient as their own baseline. Here we turn to conditional variational autoencoders to generate individualized MRI predictions given the subject's age, disease status and one previous scan. Using serial imaging data from the Alzheimer's Disease Neuroimaging Initiative, we train a novel architecture to build a latent space distribution which can be sampled from to generate future predictions of changing anatomy. This enables us to extrapolate beyond the dataset and predict MRIs up to 10 years. We evaluated the model on a held-out set from ADNI and an independent dataset (from Open Access Series of Imaging Studies). By comparing to several alternatives, we show that our model produces more individualized images with higher resolution. Further, if an individual already has a follow-up MRI, we demonstrate a usage of our model to compute a likelihood ratio classifier for disease status. In practice, the model may be able to assist in early diagnosis of AD and provide a counterfactual baseline trajectory for treatment effect estimation. Furthermore, it generates a synthetic dataset that can potentially be used for downstream tasks such as anomaly detection and classification.
△ Less
Submitted 4 August, 2024;
originally announced August 2024.
-
STAMP: Outlier-Aware Test-Time Adaptation with Stable Memory Replay
Authors:
Yongcan Yu,
Lijun Sheng,
Ran He,
Jian Liang
Abstract:
Test-time adaptation (TTA) aims to address the distribution shift between the training and test data with only unlabeled data at test time. Existing TTA methods often focus on improving recognition performance specifically for test data associated with classes in the training set. However, during the open-world inference process, there are inevitably test data instances from unknown classes, commo…
▽ More
Test-time adaptation (TTA) aims to address the distribution shift between the training and test data with only unlabeled data at test time. Existing TTA methods often focus on improving recognition performance specifically for test data associated with classes in the training set. However, during the open-world inference process, there are inevitably test data instances from unknown classes, commonly referred to as outliers. This paper pays attention to the problem that conducts both sample recognition and outlier rejection during inference while outliers exist. To address this problem, we propose a new approach called STAble Memory rePlay (STAMP), which performs optimization over a stable memory bank instead of the risky mini-batch. In particular, the memory bank is dynamically updated by selecting low-entropy and label-consistent samples in a class-balanced manner. In addition, we develop a self-weighted entropy minimization strategy that assigns higher weight to low-entropy samples. Extensive results demonstrate that STAMP outperforms existing TTA methods in terms of both recognition and outlier detection performance. The code is released at https://github.com/yuyongcan/STAMP.
△ Less
Submitted 27 August, 2024; v1 submitted 22 July, 2024;
originally announced July 2024.
-
Group Projected Subspace Pursuit for Block Sparse Signal Reconstruction: Convergence Analysis and Applications
Authors:
Roy Y. He,
Haixia Liu,
Hao Liu
Abstract:
In this paper, we present a convergence analysis of the Group Projected Subspace Pursuit (GPSP) algorithm proposed by He et al. [HKL+23] (Group Projected subspace pursuit for IDENTification of variable coefficient differential equations (GP-IDENT), Journal of Computational Physics, 494, 112526) and extend its application to general tasks of block sparse signal recovery. We prove that when the samp…
▽ More
In this paper, we present a convergence analysis of the Group Projected Subspace Pursuit (GPSP) algorithm proposed by He et al. [HKL+23] (Group Projected subspace pursuit for IDENTification of variable coefficient differential equations (GP-IDENT), Journal of Computational Physics, 494, 112526) and extend its application to general tasks of block sparse signal recovery. We prove that when the sampling matrix satisfies the Block Restricted Isometry Property (BRIP) with a sufficiently small Block Restricted Isometry Constant (BRIC), GPSP exactly recovers the true block sparse signals. When the observations are noisy, this convergence property of GPSP remains valid if the magnitude of true signal is sufficiently large. GPSP selects the features by subspace projection criterion (SPC) for candidate inclusion and response magnitude criterion (RMC) for candidate exclusion. We compare these criteria with counterparts of other state-of-the-art greedy algorithms. Our theoretical analysis and numerical ablation studies reveal that SPC is critical to the superior performances of GPSP, and that RMC can enhance the robustness of feature identification when observations contain noises. We test and compare GPSP with other methods in diverse settings, including heterogeneous random block matrices, inexact observations, face recognition, and PDE identification. We find that GPSP outperforms the other algorithms in most cases for various levels of block sparsity and block sizes, justifying its effectiveness for general applications.
△ Less
Submitted 13 July, 2024; v1 submitted 1 June, 2024;
originally announced July 2024.
-
Euler's Elastica Based Cartoon-Smooth-Texture Image Decomposition
Authors:
Roy Y. He,
Hao Liu
Abstract:
We propose a novel model for decomposing grayscale images into three distinct components: the structural part, representing sharp boundaries and regions with strong light-to-dark transitions; the smooth part, capturing soft shadows and shades; and the oscillatory part, characterizing textures and noise. To capture the homogeneous structures, we introduce a combination of $L^0$-gradient and curvatu…
▽ More
We propose a novel model for decomposing grayscale images into three distinct components: the structural part, representing sharp boundaries and regions with strong light-to-dark transitions; the smooth part, capturing soft shadows and shades; and the oscillatory part, characterizing textures and noise. To capture the homogeneous structures, we introduce a combination of $L^0$-gradient and curvature regularization on level lines. This new regularization term enforces strong sparsity on the image gradient while reducing the undesirable staircase effects as well as preserving the geometry of contours. For the smoothly varying component, we utilize the $L^2$-norm of the Laplacian that favors isotropic smoothness. To capture the oscillation, we use the inverse Sobolev seminorm. To solve the associated minimization problem, we design an efficient operator-splitting algorithm. Our algorithm effectively addresses the challenging non-convex non-smooth problem by separating it into sub-problems. Each sub-problem can be solved either directly using closed-form solutions or efficiently using the Fast Fourier Transform (FFT). We provide systematic experiments, including ablation and comparison studies, to analyze our model's behaviors and demonstrate its effectiveness as well as efficiency.
△ Less
Submitted 2 July, 2024;
originally announced July 2024.
-
MORPHEUS: Modeling Role from Personalized Dialogue History by Exploring and Utilizing Latent Space
Authors:
Yihong Tang,
Bo Wang,
Dongming Zhao,
Xiaojia Jin,
Jijun Zhang,
Ruifang He,
Yuexian Hou
Abstract:
Personalized Dialogue Generation (PDG) aims to create coherent responses according to roles or personas. Traditional PDG relies on external role data, which can be scarce and raise privacy concerns. Approaches address these issues by extracting role information from dialogue history, which often fail to generically model roles in continuous space. To overcome these limitations, we introduce a nove…
▽ More
Personalized Dialogue Generation (PDG) aims to create coherent responses according to roles or personas. Traditional PDG relies on external role data, which can be scarce and raise privacy concerns. Approaches address these issues by extracting role information from dialogue history, which often fail to generically model roles in continuous space. To overcome these limitations, we introduce a novel framework \textbf{MO}dels \textbf{R}oles from \textbf{P}ersonalized Dialogue \textbf{H}istory by \textbf{E}xploring and \textbf{U}tilizing Latent \textbf{S}pace (MORPHEUS) through a three-stage training process. Specifically, we create a persona codebook to represent roles in latent space compactly, and this codebook is used to construct a posterior distribution of role information. This method enables the model to generalize across roles, allowing the generation of personalized dialogues even for unseen roles. Experiments on both Chinese and English datasets demonstrate that MORPHEUS enhances the extraction of role information, and improves response generation without external role data. Additionally, MORPHEUS can be considered an efficient fine-tuning for large language models.
△ Less
Submitted 2 July, 2024;
originally announced July 2024.
-
Harvesting Efficient On-Demand Order Pooling from Skilled Couriers: Enhancing Graph Representation Learning for Refining Real-time Many-to-One Assignments
Authors:
Yile Liang,
Jiuxia Zhao,
Donghui Li,
Jie Feng,
Chen Zhang,
Xuetao Ding,
Jinghua Hao,
Renqing He
Abstract:
The recent past has witnessed a notable surge in on-demand food delivery (OFD) services, offering delivery fulfillment within dozens of minutes after an order is placed. In OFD, pooling multiple orders for simultaneous delivery in real-time order assignment is a pivotal efficiency source, which may in turn extend delivery time. Constructing high-quality order pooling to harmonize platform efficien…
▽ More
The recent past has witnessed a notable surge in on-demand food delivery (OFD) services, offering delivery fulfillment within dozens of minutes after an order is placed. In OFD, pooling multiple orders for simultaneous delivery in real-time order assignment is a pivotal efficiency source, which may in turn extend delivery time. Constructing high-quality order pooling to harmonize platform efficiency with the experiences of consumers and couriers, is crucial to OFD platforms. However, the complexity and real-time nature of order assignment, making extensive calculations impractical, significantly limit the potential for order consolidation. Moreover, offline environment is frequently riddled with unknown factors, posing challenges for the platform's perceptibility and pooling decisions. Nevertheless, delivery behaviors of skilled couriers (SCs) who know the environment well, can improve system awareness and effectively inform decisions. Hence a SC delivery network (SCDN) is constructed, based on an enhanced attributed heterogeneous network embedding approach tailored for OFD. It aims to extract features from rich temporal and spatial information, and uncover the latent potential for order combinations embedded within SC trajectories. Accordingly, the vast search space of order assignment can be effectively pruned through scalable similarity calculations of low-dimensional vectors, making comprehensive and high-quality pooling outcomes more easily identified in real time. SCDN has now been deployed in Meituan dispatch system. Online tests reveal that with SCDN, the pooling quality and extent have been greatly improved. And our system can boost couriers'efficiency by 45-55% during noon peak hours, while upholding the timely delivery commitment.
△ Less
Submitted 20 June, 2024;
originally announced June 2024.
-
Chumor 1.0: A Truly Funny and Challenging Chinese Humor Understanding Dataset from Ruo Zhi Ba
Authors:
Ruiqi He,
Yushu He,
Longju Bai,
Jiarui Liu,
Zhenjie Sun,
Zenghao Tang,
He Wang,
Hanchen Xia,
Naihao Deng
Abstract:
Existing humor datasets and evaluations predominantly focus on English, lacking resources for culturally nuanced humor in non-English languages like Chinese. To address this gap, we construct Chumor, a dataset sourced from Ruo Zhi Ba (RZB), a Chinese Reddit-like platform dedicated to sharing intellectually challenging and culturally specific jokes. We annotate explanations for each joke and evalua…
▽ More
Existing humor datasets and evaluations predominantly focus on English, lacking resources for culturally nuanced humor in non-English languages like Chinese. To address this gap, we construct Chumor, a dataset sourced from Ruo Zhi Ba (RZB), a Chinese Reddit-like platform dedicated to sharing intellectually challenging and culturally specific jokes. We annotate explanations for each joke and evaluate human explanations against two state-of-the-art LLMs, GPT-4o and ERNIE Bot, through A/B testing by native Chinese speakers. Our evaluation shows that Chumor is challenging even for SOTA LLMs, and the human explanations for Chumor jokes are significantly better than explanations generated by the LLMs.
△ Less
Submitted 18 June, 2024;
originally announced June 2024.
-
Trajectory Planning for Autonomous Driving in Unstructured Scenarios Based on Graph Neural Network and Numerical Optimization
Authors:
Sumin Zhang,
Kuo Li,
Rui He,
Zhiwei Meng,
Yupeng Chang,
Xiaosong Jin,
Ri Bai
Abstract:
In unstructured environments, obstacles are diverse and lack lane markings, making trajectory planning for intelligent vehicles a challenging task. Traditional trajectory planning methods typically involve multiple stages, including path planning, speed planning, and trajectory optimization. These methods require the manual design of numerous parameters for each stage, resulting in significant wor…
▽ More
In unstructured environments, obstacles are diverse and lack lane markings, making trajectory planning for intelligent vehicles a challenging task. Traditional trajectory planning methods typically involve multiple stages, including path planning, speed planning, and trajectory optimization. These methods require the manual design of numerous parameters for each stage, resulting in significant workload and computational burden. While end-to-end trajectory planning methods are simple and efficient, they often fail to ensure that the trajectory meets vehicle dynamics and obstacle avoidance constraints in unstructured scenarios. Therefore, this paper proposes a novel trajectory planning method based on Graph Neural Networks (GNN) and numerical optimization. The proposed method consists of two stages: (1) initial trajectory prediction using the GNN, (2) trajectory optimization using numerical optimization. First, the graph neural network processes the environment information and predicts a rough trajectory, replacing traditional path and speed planning. This predicted trajectory serves as the initial solution for the numerical optimization stage, which optimizes the trajectory to ensure compliance with vehicle dynamics and obstacle avoidance constraints. We conducted simulation experiments to validate the feasibility of the proposed algorithm and compared it with other mainstream planning algorithms. The results demonstrate that the proposed method simplifies the trajectory planning process and significantly improves planning efficiency.
△ Less
Submitted 13 June, 2024;
originally announced June 2024.
-
ZeroSmooth: Training-free Diffuser Adaptation for High Frame Rate Video Generation
Authors:
Shaoshu Yang,
Yong Zhang,
Xiaodong Cun,
Ying Shan,
Ran He
Abstract:
Video generation has made remarkable progress in recent years, especially since the advent of the video diffusion models. Many video generation models can produce plausible synthetic videos, e.g., Stable Video Diffusion (SVD). However, most video models can only generate low frame rate videos due to the limited GPU memory as well as the difficulty of modeling a large set of frames. The training vi…
▽ More
Video generation has made remarkable progress in recent years, especially since the advent of the video diffusion models. Many video generation models can produce plausible synthetic videos, e.g., Stable Video Diffusion (SVD). However, most video models can only generate low frame rate videos due to the limited GPU memory as well as the difficulty of modeling a large set of frames. The training videos are always uniformly sampled at a specified interval for temporal compression. Previous methods promote the frame rate by either training a video interpolation model in pixel space as a postprocessing stage or training an interpolation model in latent space for a specific base video model. In this paper, we propose a training-free video interpolation method for generative video diffusion models, which is generalizable to different models in a plug-and-play manner. We investigate the non-linearity in the feature space of video diffusion models and transform a video model into a self-cascaded video diffusion model with incorporating the designed hidden state correction modules. The self-cascaded architecture and the correction module are proposed to retain the temporal consistency between key frames and the interpolated frames. Extensive evaluations are preformed on multiple popular video models to demonstrate the effectiveness of the propose method, especially that our training-free method is even comparable to trained interpolation models supported by huge compute resources and large-scale datasets.
△ Less
Submitted 2 June, 2024;
originally announced June 2024.
-
A Point-Neighborhood Learning Framework for Nasal Endoscope Image Segmentation
Authors:
Pengyu Jie,
Wanquan Liu,
Chenqiang Gao,
Yihui Wen,
Rui He,
Pengcheng Li,
Jintao Zhang,
Deyu Meng
Abstract:
The lesion segmentation on endoscopic images is challenging due to its complex and ambiguous features. Fully-supervised deep learning segmentation methods can receive good performance based on entirely pixel-level labeled dataset but greatly increase experts' labeling burden. Semi-supervised and weakly supervised methods can ease labeling burden, but heavily strengthen the learning difficulty. To…
▽ More
The lesion segmentation on endoscopic images is challenging due to its complex and ambiguous features. Fully-supervised deep learning segmentation methods can receive good performance based on entirely pixel-level labeled dataset but greatly increase experts' labeling burden. Semi-supervised and weakly supervised methods can ease labeling burden, but heavily strengthen the learning difficulty. To alleviate this difficulty, weakly semi-supervised segmentation adopts a new annotation protocol of adding a large number of point annotation samples into a few pixel-level annotation samples. However, existing methods only mine points' limited information while ignoring reliable prior surrounding the point annotations. In this paper, we propose a weakly semi-supervised method called Point-Neighborhood Learning (PNL) framework. To mine the prior of the pixels surrounding the annotated point, we transform a single-point annotation into a circular area named a point-neighborhood. We propose point-neighborhood supervision loss and pseudo-label scoring mechanism to enhance training supervision. Point-neighborhoods are also used to augment the data diversity. Our method greatly improves performance without changing the structure of segmentation network. Comprehensive experiments show the superiority of our method over the other existing methods, demonstrating its effectiveness in point-annotated medical images. The project code will be available on: https://github.com/ParryJay/PNL.
△ Less
Submitted 30 May, 2024;
originally announced May 2024.
-
Visual Anchors Are Strong Information Aggregators For Multimodal Large Language Model
Authors:
Haogeng Liu,
Quanzeng You,
Xiaotian Han,
Yongfei Liu,
Huaibo Huang,
Ran He,
Hongxia Yang
Abstract:
In the realm of Multimodal Large Language Models (MLLMs), vision-language connector plays a crucial role to link the pre-trained vision encoders with Large Language Models (LLMs). Despite its importance, the vision-language connector has been relatively less explored. In this study, we aim to propose a strong vision-language connector that enables MLLMs to achieve high accuracy while maintain low…
▽ More
In the realm of Multimodal Large Language Models (MLLMs), vision-language connector plays a crucial role to link the pre-trained vision encoders with Large Language Models (LLMs). Despite its importance, the vision-language connector has been relatively less explored. In this study, we aim to propose a strong vision-language connector that enables MLLMs to achieve high accuracy while maintain low computation cost. We first reveal the existence of the visual anchors in Vision Transformer and propose a cost-effective search algorithm to extract them. Building on these findings, we introduce the Anchor Former (AcFormer), a novel vision-language connector designed to leverage the rich prior knowledge obtained from these visual anchors during pretraining, guiding the aggregation of information. Through extensive experimentation, we demonstrate that the proposed method significantly reduces computational costs by nearly two-thirds compared with baseline, while simultaneously outperforming baseline methods. This highlights the effectiveness and efficiency of AcFormer.
△ Less
Submitted 28 May, 2024;
originally announced May 2024.
-
Analytic Federated Learning
Authors:
Huiping Zhuang,
Run He,
Kai Tong,
Di Fang,
Han Sun,
Haoran Li,
Tianyi Chen,
Ziqian Zeng
Abstract:
In this paper, we introduce analytic federated learning (AFL), a new training paradigm that brings analytical (i.e., closed-form) solutions to the federated learning (FL) community. Our AFL draws inspiration from analytic learning -- a gradient-free technique that trains neural networks with analytical solutions in one epoch. In the local client training stage, the AFL facilitates a one-epoch trai…
▽ More
In this paper, we introduce analytic federated learning (AFL), a new training paradigm that brings analytical (i.e., closed-form) solutions to the federated learning (FL) community. Our AFL draws inspiration from analytic learning -- a gradient-free technique that trains neural networks with analytical solutions in one epoch. In the local client training stage, the AFL facilitates a one-epoch training, eliminating the necessity for multi-epoch updates. In the aggregation stage, we derive an absolute aggregation (AA) law. This AA law allows a single-round aggregation, removing the need for multiple aggregation rounds. More importantly, the AFL exhibits a \textit{weight-invariant} property, meaning that regardless of how the full dataset is distributed among clients, the aggregated result remains identical. This could spawn various potentials, such as data heterogeneity invariance, client-number invariance, absolute convergence, and being hyperparameter-free (our AFL is the first hyperparameter-free method in FL history). We conduct experiments across various FL settings including extremely non-IID ones, and scenarios with a large number of clients (e.g., $\ge 1000$). In all these settings, our AFL constantly performs competitively while existing FL techniques encounter various obstacles. Code is available at \url{https://github.com/ZHUANGHP/Analytic-federated-learning}
△ Less
Submitted 25 May, 2024;
originally announced May 2024.
-
Diverse Teacher-Students for Deep Safe Semi-Supervised Learning under Class Mismatch
Authors:
Qikai Wang,
Rundong He,
Yongshun Gong,
Chunxiao Ren,
Haoliang Sun,
Xiaoshui Huang,
Yilong Yin
Abstract:
Semi-supervised learning can significantly boost model performance by leveraging unlabeled data, particularly when labeled data is scarce. However, real-world unlabeled data often contain unseen-class samples, which can hinder the classification of seen classes. To address this issue, mainstream safe SSL methods suggest detecting and discarding unseen-class samples from unlabeled data. Nevertheles…
▽ More
Semi-supervised learning can significantly boost model performance by leveraging unlabeled data, particularly when labeled data is scarce. However, real-world unlabeled data often contain unseen-class samples, which can hinder the classification of seen classes. To address this issue, mainstream safe SSL methods suggest detecting and discarding unseen-class samples from unlabeled data. Nevertheless, these methods typically employ a single-model strategy to simultaneously tackle both the classification of seen classes and the detection of unseen classes. Our research indicates that such an approach may lead to conflicts during training, resulting in suboptimal model optimization. Inspired by this, we introduce a novel framework named Diverse Teacher-Students (\textbf{DTS}), which uniquely utilizes dual teacher-student models to individually and effectively handle these two tasks. DTS employs a novel uncertainty score to softly separate unseen-class and seen-class data from the unlabeled set, and intelligently creates an additional ($K$+1)-th class supervisory signal for training. By training both teacher-student models with all unlabeled samples, DTS can enhance the classification of seen classes while simultaneously improving the detection of unseen classes. Comprehensive experiments demonstrate that DTS surpasses baseline methods across a variety of datasets and configurations. Our code and models can be publicly accessible on the link https://github.com/Zhanlo/DTS.
△ Less
Submitted 25 May, 2024;
originally announced May 2024.
-
PitVQA: Image-grounded Text Embedding LLM for Visual Question Answering in Pituitary Surgery
Authors:
Runlong He,
Mengya Xu,
Adrito Das,
Danyal Z. Khan,
Sophia Bano,
Hani J. Marcus,
Danail Stoyanov,
Matthew J. Clarkson,
Mobarakol Islam
Abstract:
Visual Question Answering (VQA) within the surgical domain, utilizing Large Language Models (LLMs), offers a distinct opportunity to improve intra-operative decision-making and facilitate intuitive surgeon-AI interaction. However, the development of LLMs for surgical VQA is hindered by the scarcity of diverse and extensive datasets with complex reasoning tasks. Moreover, contextual fusion of the i…
▽ More
Visual Question Answering (VQA) within the surgical domain, utilizing Large Language Models (LLMs), offers a distinct opportunity to improve intra-operative decision-making and facilitate intuitive surgeon-AI interaction. However, the development of LLMs for surgical VQA is hindered by the scarcity of diverse and extensive datasets with complex reasoning tasks. Moreover, contextual fusion of the image and text modalities remains an open research challenge due to the inherent differences between these two types of information and the complexity involved in aligning them. This paper introduces PitVQA, a novel dataset specifically designed for VQA in endonasal pituitary surgery and PitVQA-Net, an adaptation of the GPT2 with a novel image-grounded text embedding for surgical VQA. PitVQA comprises 25 procedural videos and a rich collection of question-answer pairs spanning crucial surgical aspects such as phase and step recognition, context understanding, tool detection and localization, and tool-tissue interactions. PitVQA-Net consists of a novel image-grounded text embedding that projects image and text features into a shared embedding space and GPT2 Backbone with an excitation block classification head to generate contextually relevant answers within the complex domain of endonasal pituitary surgery. Our image-grounded text embedding leverages joint embedding, cross-attention and contextual representation to understand the contextual relationship between questions and surgical images. We demonstrate the effectiveness of PitVQA-Net on both the PitVQA and the publicly available EndoVis18-VQA dataset, achieving improvements in balanced accuracy of 8% and 9% over the most recent baselines, respectively. Our code and dataset is available at https://github.com/mobarakol/PitVQA.
△ Less
Submitted 22 May, 2024;
originally announced May 2024.
-
Semantic Equitable Clustering: A Simple, Fast and Effective Strategy for Vision Transformer
Authors:
Qihang Fan,
Huaibo Huang,
Mingrui Chen,
Ran He
Abstract:
The Vision Transformer (ViT) has gained prominence for its superior relational modeling prowess. However, its global attention mechanism's quadratic complexity poses substantial computational burdens. A common remedy spatially groups tokens for self-attention, reducing computational requirements. Nonetheless, this strategy neglects semantic information in tokens, possibly scattering semantically-l…
▽ More
The Vision Transformer (ViT) has gained prominence for its superior relational modeling prowess. However, its global attention mechanism's quadratic complexity poses substantial computational burdens. A common remedy spatially groups tokens for self-attention, reducing computational requirements. Nonetheless, this strategy neglects semantic information in tokens, possibly scattering semantically-linked tokens across distinct groups, thus compromising the efficacy of self-attention intended for modeling inter-token dependencies. Motivated by these insights, we introduce a fast and balanced clustering method, named \textbf{S}emantic \textbf{E}quitable \textbf{C}lustering (SEC). SEC clusters tokens based on their global semantic relevance in an efficient, straightforward manner. In contrast to traditional clustering methods requiring multiple iterations, our method achieves token clustering in a single pass. Additionally, SEC regulates the number of tokens per cluster, ensuring a balanced distribution for effective parallel processing on current computational platforms without necessitating further optimization. Capitalizing on SEC, we propose a versatile vision backbone, SecViT. Comprehensive experiments in image classification, object detection, instance segmentation, and semantic segmentation validate to the effectiveness of SecViT. Remarkably, SecViT attains an impressive \textbf{84.2\%} image classification accuracy with only \textbf{27M} parameters and \textbf{4.4G} FLOPs, without the need for for additional supervision or data. Code will be available at \url{https://github.com/qhfan/SecViT}.
△ Less
Submitted 22 May, 2024;
originally announced May 2024.
-
Vision Transformer with Sparse Scan Prior
Authors:
Qihang Fan,
Huaibo Huang,
Mingrui Chen,
Ran He
Abstract:
In recent years, Transformers have achieved remarkable progress in computer vision tasks. However, their global modeling often comes with substantial computational overhead, in stark contrast to the human eye's efficient information processing. Inspired by the human eye's sparse scanning mechanism, we propose a \textbf{S}parse \textbf{S}can \textbf{S}elf-\textbf{A}ttention mechanism (…
▽ More
In recent years, Transformers have achieved remarkable progress in computer vision tasks. However, their global modeling often comes with substantial computational overhead, in stark contrast to the human eye's efficient information processing. Inspired by the human eye's sparse scanning mechanism, we propose a \textbf{S}parse \textbf{S}can \textbf{S}elf-\textbf{A}ttention mechanism ($\rm{S}^3\rm{A}$). This mechanism predefines a series of Anchors of Interest for each token and employs local attention to efficiently model the spatial information around these anchors, avoiding redundant global modeling and excessive focus on local information. This approach mirrors the human eye's functionality and significantly reduces the computational load of vision models. Building on $\rm{S}^3\rm{A}$, we introduce the \textbf{S}parse \textbf{S}can \textbf{Vi}sion \textbf{T}ransformer (SSViT). Extensive experiments demonstrate the outstanding performance of SSViT across a variety of tasks. Specifically, on ImageNet classification, without additional supervision or training data, SSViT achieves top-1 accuracies of \textbf{84.4\%/85.7\%} with \textbf{4.4G/18.2G} FLOPs. SSViT also excels in downstream tasks such as object detection, instance segmentation, and semantic segmentation. Its robustness is further validated across diverse datasets. Code will be available at \url{https://github.com/qhfan/SSViT}.
△ Less
Submitted 22 May, 2024;
originally announced May 2024.
-
Revealing the value of Repository Centrality in lifespan prediction of Open Source Software Projects
Authors:
Runzhi He,
Hengzhi Ye,
Minghui Zhou
Abstract:
Background: Open Source Software is the building block of modern software. However, the prevalence of project deprecation in the open source world weakens the integrity of the downstream systems and the broad ecosystem. Therefore it calls for efforts in monitoring and predicting project deprecations, empowering stakeholders to take proactive measures. Challenge: Existing techniques mainly focus on…
▽ More
Background: Open Source Software is the building block of modern software. However, the prevalence of project deprecation in the open source world weakens the integrity of the downstream systems and the broad ecosystem. Therefore it calls for efforts in monitoring and predicting project deprecations, empowering stakeholders to take proactive measures. Challenge: Existing techniques mainly focus on static features on a point in time to make predictions, resulting in limited effects. Goal: We propose a novel metric from the user-repository network, and leverage the metric to fit project deprecation predictors and prove its real-life implications. Method: We establish a comprehensive dataset containing 103,354 non-fork GitHub OSS projects spanning from 2011 to 2023. We propose repository centrality, a family of HITS weights that captures shifts in the popularity of a repository in the repository-user star network. Further with the metric, we utilize the advancements in gradient boosting and deep learning to fit survival analysis models to predict project lifespan or its survival hazard. Results: Our study reveals a correlation between the HITS centrality metrics and the repository deprecation risk. A drop in the HITS weights of a repository indicates a decline in its centrality and prevalence, leading to an increase in its deprecation risk and a decrease in its expected lifespan. Our predictive models powered by repository centrality and other repository features achieve satisfactory accuracy on the test set, with repository centrality being the most significant feature among all. Implications: This research offers a novel perspective on understanding the effect of prevalence on the deprecation of OSS repositories. Our approach to predict repository deprecation help detect health status of project and take actions in advance, fostering a more resilient OSS ecosystem.
△ Less
Submitted 13 May, 2024;
originally announced May 2024.
-
Generative Diffusion Model (GDM) for Optimization of Wi-Fi Networks
Authors:
Tie Liu,
Xuming Fang,
Rong He
Abstract:
Generative Diffusion Models (GDMs), have made significant strides in modeling complex data distributions across diverse domains. Meanwhile, Deep Reinforcement Learning (DRL) has demonstrated substantial improvements in optimizing Wi-Fi network performance. Wi-Fi optimization problems are highly challenging to model mathematically, and DRL methods can bypass complex mathematical modeling, while GDM…
▽ More
Generative Diffusion Models (GDMs), have made significant strides in modeling complex data distributions across diverse domains. Meanwhile, Deep Reinforcement Learning (DRL) has demonstrated substantial improvements in optimizing Wi-Fi network performance. Wi-Fi optimization problems are highly challenging to model mathematically, and DRL methods can bypass complex mathematical modeling, while GDMs excel in handling complex data modeling. Therefore, combining DRL with GDMs can mutually enhance their capabilities. The current MAC layer access mechanism in Wi-Fi networks is the Distributed Coordination Function (DCF), which dramatically declines in performance with a high number of terminals. In this paper, we apply diffusion models to deep deterministic policy gradient, namely the Deep Diffusion Deterministic Policy (D3PG) algorithm to optimize the Wi-Fi performance. Although such integrations have been explored previously, we are the first to apply it to Wi-Fi network performance optimization. We propose an access mechanism that jointly adjusts the contention window and frame length based on the D3PG algorithm. Through simulations, we have demonstrated that this mechanism significantly outperforms existing Wi-Fi standards in dense Wi-Fi scenarios, maintaining performance even as the number of users sharply increases.
△ Less
Submitted 24 April, 2024;
originally announced April 2024.
-
Band-Attention Modulated RetNet for Face Forgery Detection
Authors:
Zhida Zhang,
Jie Cao,
Wenkui Yang,
Qihang Fan,
Kai Zhou,
Ran He
Abstract:
The transformer networks are extensively utilized in face forgery detection due to their scalability across large datasets.Despite their success, transformers face challenges in balancing the capture of global context, which is crucial for unveiling forgery clues, with computational complexity.To mitigate this issue, we introduce Band-Attention modulated RetNet (BAR-Net), a lightweight network des…
▽ More
The transformer networks are extensively utilized in face forgery detection due to their scalability across large datasets.Despite their success, transformers face challenges in balancing the capture of global context, which is crucial for unveiling forgery clues, with computational complexity.To mitigate this issue, we introduce Band-Attention modulated RetNet (BAR-Net), a lightweight network designed to efficiently process extensive visual contexts while avoiding catastrophic forgetting.Our approach empowers the target token to perceive global information by assigning differential attention levels to tokens at varying distances. We implement self-attention along both spatial axes, thereby maintaining spatial priors and easing the computational burden.Moreover, we present the adaptive frequency Band-Attention Modulation mechanism, which treats the entire Discrete Cosine Transform spectrogram as a series of frequency bands with learnable weights.Together, BAR-Net achieves favorable performance on several face forgery datasets, outperforming current state-of-the-art methods.
△ Less
Submitted 1 July, 2024; v1 submitted 9 April, 2024;
originally announced April 2024.
-
SportsHHI: A Dataset for Human-Human Interaction Detection in Sports Videos
Authors:
Tao Wu,
Runyu He,
Gangshan Wu,
Limin Wang
Abstract:
Video-based visual relation detection tasks, such as video scene graph generation, play important roles in fine-grained video understanding. However, current video visual relation detection datasets have two main limitations that hinder the progress of research in this area. First, they do not explore complex human-human interactions in multi-person scenarios. Second, the relation types of existin…
▽ More
Video-based visual relation detection tasks, such as video scene graph generation, play important roles in fine-grained video understanding. However, current video visual relation detection datasets have two main limitations that hinder the progress of research in this area. First, they do not explore complex human-human interactions in multi-person scenarios. Second, the relation types of existing datasets have relatively low-level semantics and can be often recognized by appearance or simple prior information, without the need for detailed spatio-temporal context reasoning. Nevertheless, comprehending high-level interactions between humans is crucial for understanding complex multi-person videos, such as sports and surveillance videos. To address this issue, we propose a new video visual relation detection task: video human-human interaction detection, and build a dataset named SportsHHI for it. SportsHHI contains 34 high-level interaction classes from basketball and volleyball sports. 118,075 human bounding boxes and 50,649 interaction instances are annotated on 11,398 keyframes. To benchmark this, we propose a two-stage baseline method and conduct extensive experiments to reveal the key factors for a successful human-human interaction detector. We hope that SportsHHI can stimulate research on human interaction understanding in videos and promote the development of spatio-temporal context modeling techniques in video visual relation detection.
△ Less
Submitted 6 April, 2024;
originally announced April 2024.
-
CLIP-driven Outliers Synthesis for few-shot OOD detection
Authors:
Hao Sun,
Rundong He,
Zhongyi Han,
Zhicong Lin,
Yongshun Gong,
Yilong Yin
Abstract:
Few-shot OOD detection focuses on recognizing out-of-distribution (OOD) images that belong to classes unseen during training, with the use of only a small number of labeled in-distribution (ID) images. Up to now, a mainstream strategy is based on large-scale vision-language models, such as CLIP. However, these methods overlook a crucial issue: the lack of reliable OOD supervision information, whic…
▽ More
Few-shot OOD detection focuses on recognizing out-of-distribution (OOD) images that belong to classes unseen during training, with the use of only a small number of labeled in-distribution (ID) images. Up to now, a mainstream strategy is based on large-scale vision-language models, such as CLIP. However, these methods overlook a crucial issue: the lack of reliable OOD supervision information, which can lead to biased boundaries between in-distribution (ID) and OOD. To tackle this problem, we propose CLIP-driven Outliers Synthesis~(CLIP-OS). Firstly, CLIP-OS enhances patch-level features' perception by newly proposed patch uniform convolution, and adaptively obtains the proportion of ID-relevant information by employing CLIP-surgery-discrepancy, thus achieving separation between ID-relevant and ID-irrelevant. Next, CLIP-OS synthesizes reliable OOD data by mixing up ID-relevant features from different classes to provide OOD supervision information. Afterward, CLIP-OS leverages synthetic OOD samples by unknown-aware prompt learning to enhance the separability of ID and OOD. Extensive experiments across multiple benchmarks demonstrate that CLIP-OS achieves superior few-shot OOD detection capability.
△ Less
Submitted 30 March, 2024;
originally announced April 2024.
-
ViTAR: Vision Transformer with Any Resolution
Authors:
Qihang Fan,
Quanzeng You,
Xiaotian Han,
Yongfei Liu,
Yunzhe Tao,
Huaibo Huang,
Ran He,
Hongxia Yang
Abstract:
This paper tackles a significant challenge faced by Vision Transformers (ViTs): their constrained scalability across different image resolutions. Typically, ViTs experience a performance decline when processing resolutions different from those seen during training. Our work introduces two key innovations to address this issue. Firstly, we propose a novel module for dynamic resolution adjustment, d…
▽ More
This paper tackles a significant challenge faced by Vision Transformers (ViTs): their constrained scalability across different image resolutions. Typically, ViTs experience a performance decline when processing resolutions different from those seen during training. Our work introduces two key innovations to address this issue. Firstly, we propose a novel module for dynamic resolution adjustment, designed with a single Transformer block, specifically to achieve highly efficient incremental token integration. Secondly, we introduce fuzzy positional encoding in the Vision Transformer to provide consistent positional awareness across multiple resolutions, thereby preventing overfitting to any single training resolution. Our resulting model, ViTAR (Vision Transformer with Any Resolution), demonstrates impressive adaptability, achieving 83.3\% top-1 accuracy at a 1120x1120 resolution and 80.4\% accuracy at a 4032x4032 resolution, all while reducing computational costs. ViTAR also shows strong performance in downstream tasks such as instance and semantic segmentation and can easily combined with self-supervised learning techniques like Masked AutoEncoder. Our work provides a cost-effective solution for enhancing the resolution scalability of ViTs, paving the way for more versatile and efficient high-resolution image processing.
△ Less
Submitted 28 March, 2024; v1 submitted 27 March, 2024;
originally announced March 2024.
-
MUTE-SLAM: Real-Time Neural SLAM with Multiple Tri-Plane Hash Representations
Authors:
Yifan Yan,
Ruomin He,
Zhenghua Liu
Abstract:
We introduce MUTE-SLAM, a real-time neural RGB-D SLAM system employing multiple tri-plane hash-encodings for efficient scene representation. MUTE-SLAM effectively tracks camera positions and incrementally builds a scalable multi-map representation for both small and large indoor environments. As previous methods often require pre-defined scene boundaries, MUTE-SLAM dynamically allocates sub-maps f…
▽ More
We introduce MUTE-SLAM, a real-time neural RGB-D SLAM system employing multiple tri-plane hash-encodings for efficient scene representation. MUTE-SLAM effectively tracks camera positions and incrementally builds a scalable multi-map representation for both small and large indoor environments. As previous methods often require pre-defined scene boundaries, MUTE-SLAM dynamically allocates sub-maps for newly observed local regions, enabling constraint-free mapping without prior scene information. Unlike traditional grid-based methods, we use three orthogonal axis-aligned planes for hash-encoding scene properties, significantly reducing hash collisions and the number of trainable parameters. This hybrid approach not only ensures real-time performance but also enhances the fidelity of surface reconstruction. Furthermore, our optimization strategy concurrently optimizes all sub-maps intersecting with the current camera frustum, ensuring global consistency. Extensive testing on both real-world and synthetic datasets has shown that MUTE-SLAM delivers state-of-the-art surface reconstruction quality and competitive tracking performance across diverse indoor settings. The code will be made public upon acceptance of the paper.
△ Less
Submitted 7 July, 2024; v1 submitted 26 March, 2024;
originally announced March 2024.
-
DS-AL: A Dual-Stream Analytic Learning for Exemplar-Free Class-Incremental Learning
Authors:
Huiping Zhuang,
Run He,
Kai Tong,
Ziqian Zeng,
Cen Chen,
Zhiping Lin
Abstract:
Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic forgetting, far more so than replay-based techniques that retain access to past samples. In this paper, to solve the exemplar-free CIL problem, we propose a Dual-Stream Analytic Learning (DS-AL) approach. The DS-AL contains…
▽ More
Class-incremental learning (CIL) under an exemplar-free constraint has presented a significant challenge. Existing methods adhering to this constraint are prone to catastrophic forgetting, far more so than replay-based techniques that retain access to past samples. In this paper, to solve the exemplar-free CIL problem, we propose a Dual-Stream Analytic Learning (DS-AL) approach. The DS-AL contains a main stream offering an analytical (i.e., closed-form) linear solution, and a compensation stream improving the inherent under-fitting limitation due to adopting linear mapping. The main stream redefines the CIL problem into a Concatenated Recursive Least Squares (C-RLS) task, allowing an equivalence between the CIL and its joint-learning counterpart. The compensation stream is governed by a Dual-Activation Compensation (DAC) module. This module re-activates the embedding with a different activation function from the main stream one, and seeks fitting compensation by projecting the embedding to the null space of the main stream's linear mapping. Empirical results demonstrate that the DS-AL, despite being an exemplar-free technique, delivers performance comparable with or better than that of replay-based methods across various datasets, including CIFAR-100, ImageNet-100 and ImageNet-Full. Additionally, the C-RLS' equivalent property allows the DS-AL to execute CIL in a phase-invariant manner. This is evidenced by a never-before-seen 500-phase CIL ImageNet task, which performs on a level identical to a 5-phase one. Our codes are available at https://github.com/ZHUANGHP/Analytic-continual-learning.
△ Less
Submitted 26 March, 2024;
originally announced March 2024.
-
AOCIL: Exemplar-free Analytic Online Class Incremental Learning with Low Time and Resource Consumption
Authors:
Huiping Zhuang,
Yuchen Liu,
Run He,
Kai Tong,
Ziqian Zeng,
Cen Chen,
Yi Wang,
Lap-Pui Chau
Abstract:
Online Class Incremental Learning (OCIL) aims to train the model in a task-by-task manner, where data arrive in mini-batches at a time while previous data are not accessible. A significant challenge is known as Catastrophic Forgetting, i.e., loss of the previous knowledge on old data. To address this, replay-based methods show competitive results but invade data privacy, while exemplar-free method…
▽ More
Online Class Incremental Learning (OCIL) aims to train the model in a task-by-task manner, where data arrive in mini-batches at a time while previous data are not accessible. A significant challenge is known as Catastrophic Forgetting, i.e., loss of the previous knowledge on old data. To address this, replay-based methods show competitive results but invade data privacy, while exemplar-free methods protect data privacy but struggle for accuracy. In this paper, we proposed an exemplar-free approach -- Analytic Online Class Incremental Learning (AOCIL). Instead of back-propagation, we design the Analytic Classifier (AC) updated by recursive least square, cooperating with a frozen backbone. AOCIL simultaneously achieves high accuracy, low resource consumption and data privacy protection. We conduct massive experiments on four existing benchmark datasets, and the results demonstrate the strong capability of handling OCIL scenarios. Codes will be ready.
△ Less
Submitted 23 March, 2024;
originally announced March 2024.
-
G-ACIL: Analytic Learning for Exemplar-Free Generalized Class Incremental Learning
Authors:
Huiping Zhuang,
Yizhu Chen,
Di Fang,
Run He,
Kai Tong,
Hongxin Wei,
Ziqian Zeng,
Cen Chen
Abstract:
Class incremental learning (CIL) trains a network on sequential tasks with separated categories but suffers from catastrophic forgetting, where models quickly lose previously learned knowledge when acquiring new tasks. The generalized CIL (GCIL) aims to address the CIL problem in a more real-world scenario, where incoming data have mixed data categories and unknown sample size distribution, leadin…
▽ More
Class incremental learning (CIL) trains a network on sequential tasks with separated categories but suffers from catastrophic forgetting, where models quickly lose previously learned knowledge when acquiring new tasks. The generalized CIL (GCIL) aims to address the CIL problem in a more real-world scenario, where incoming data have mixed data categories and unknown sample size distribution, leading to intensified forgetting. Existing attempts for the GCIL either have poor performance, or invade data privacy by saving historical exemplars. To address this, in this paper, we propose an exemplar-free generalized analytic class incremental learning (G-ACIL). The G-ACIL adopts analytic learning (a gradient-free training technique), and delivers an analytical solution (i.e., closed-form) to the GCIL scenario. This solution is derived via decomposing the incoming data into exposed and unexposed classes, allowing an equivalence between the incremental learning and its joint training, i.e., the weight-invariant property. Such an equivalence is theoretically validated through matrix analysis tools, and hence contributes interpretability in GCIL. It is also empirically evidenced by experiments on various datasets and settings of GCIL. The results show that the G-ACIL exhibits leading performance with high robustness compared with existing competitive GCIL methods. Codes will be ready at \url{https://github.com/ZHUANGHP/Analytic-continual-learning}.
△ Less
Submitted 13 April, 2024; v1 submitted 22 March, 2024;
originally announced March 2024.
-
Learning from Models and Data for Visual Grounding
Authors:
Ruozhen He,
Paola Cascante-Bonilla,
Ziyan Yang,
Alexander C. Berg,
Vicente Ordonez
Abstract:
We introduce SynGround, a novel framework that combines data-driven learning and knowledge transfer from various large-scale pretrained models to enhance the visual grounding capabilities of a pretrained vision-and-language model. The knowledge transfer from the models initiates the generation of image descriptions through an image description generator. These descriptions serve dual purposes: the…
▽ More
We introduce SynGround, a novel framework that combines data-driven learning and knowledge transfer from various large-scale pretrained models to enhance the visual grounding capabilities of a pretrained vision-and-language model. The knowledge transfer from the models initiates the generation of image descriptions through an image description generator. These descriptions serve dual purposes: they act as prompts for synthesizing images through a text-to-image generator, and as queries for synthesizing text, from which phrases are extracted using a large language model. Finally, we leverage an open-vocabulary object detector to generate synthetic bounding boxes for the synthetic images and texts. We finetune a pretrained vision-and-language model on this dataset by optimizing a mask-attention consistency objective that aligns region annotations with gradient-based model explanations. The resulting model improves the grounding capabilities of an off-the-shelf vision-and-language model. Particularly, SynGround improves the pointing game accuracy of ALBEF on the Flickr30k dataset from 79.38% to 87.26%, and on RefCOCO+ Test A from 69.35% to 79.06% and on RefCOCO+ Test B from 53.77% to 63.67%.
△ Less
Submitted 20 March, 2024;
originally announced March 2024.
-
REAL: Representation Enhanced Analytic Learning for Exemplar-free Class-incremental Learning
Authors:
Run He,
Huiping Zhuang,
Di Fang,
Yizhu Chen,
Kai Tong,
Cen Chen
Abstract:
Exemplar-free class-incremental learning (EFCIL) aims to mitigate catastrophic forgetting in class-incremental learning without available historical data. Compared with its counterpart (replay-based CIL) that stores historical samples, the EFCIL suffers more from forgetting issues under the exemplar-free constraint. In this paper, inspired by the recently developed analytic learning (AL) based CIL…
▽ More
Exemplar-free class-incremental learning (EFCIL) aims to mitigate catastrophic forgetting in class-incremental learning without available historical data. Compared with its counterpart (replay-based CIL) that stores historical samples, the EFCIL suffers more from forgetting issues under the exemplar-free constraint. In this paper, inspired by the recently developed analytic learning (AL) based CIL, we propose a representation enhanced analytic learning (REAL) for EFCIL. The REAL constructs a dual-stream base pretraining (DS-BPT) and a representation enhancing distillation (RED) process to enhance the representation of the extractor. The DS-BPT pretrains model in streams of both supervised learning and self-supervised contrastive learning (SSCL) for base knowledge extraction. The RED process distills the supervised knowledge to the SSCL pretrained backbone and facilitates a subsequent AL-basd CIL that converts the CIL to a recursive least-square problem. Our method addresses the issue of insufficient discriminability in representations of unseen data caused by a frozen backbone in the existing AL-based CIL. Empirical results on various datasets including CIFAR-100, ImageNet-100 and ImageNet-1k, demonstrate that our REAL outperforms the state-of-the-arts in EFCIL, and achieves comparable or even more superior performance compared with the replay-based methods.
△ Less
Submitted 20 March, 2024;
originally announced March 2024.
-
DiffMAC: Diffusion Manifold Hallucination Correction for High Generalization Blind Face Restoration
Authors:
Nan Gao,
Jia Li,
Huaibo Huang,
Zhi Zeng,
Ke Shang,
Shuwu Zhang,
Ran He
Abstract:
Blind face restoration (BFR) is a highly challenging problem due to the uncertainty of degradation patterns. Current methods have low generalization across photorealistic and heterogeneous domains. In this paper, we propose a Diffusion-Information-Diffusion (DID) framework to tackle diffusion manifold hallucination correction (DiffMAC), which achieves high-generalization face restoration in divers…
▽ More
Blind face restoration (BFR) is a highly challenging problem due to the uncertainty of degradation patterns. Current methods have low generalization across photorealistic and heterogeneous domains. In this paper, we propose a Diffusion-Information-Diffusion (DID) framework to tackle diffusion manifold hallucination correction (DiffMAC), which achieves high-generalization face restoration in diverse degraded scenes and heterogeneous domains. Specifically, the first diffusion stage aligns the restored face with spatial feature embedding of the low-quality face based on AdaIN, which synthesizes degradation-removal results but with uncontrollable artifacts for some hard cases. Based on Stage I, Stage II considers information compression using manifold information bottleneck (MIB) and finetunes the first diffusion model to improve facial fidelity. DiffMAC effectively fights against blind degradation patterns and synthesizes high-quality faces with attribute and identity consistencies. Experimental results demonstrate the superiority of DiffMAC over state-of-the-art methods, with a high degree of generalization in real-world and heterogeneous settings. The source code and models will be public.
△ Less
Submitted 15 March, 2024;
originally announced March 2024.
-
CSCNET: Class-Specified Cascaded Network for Compositional Zero-Shot Learning
Authors:
Yanyi Zhang,
Qi Jia,
Xin Fan,
Yu Liu,
Ran He
Abstract:
Attribute and object (A-O) disentanglement is a fundamental and critical problem for Compositional Zero-shot Learning (CZSL), whose aim is to recognize novel A-O compositions based on foregone knowledge. Existing methods based on disentangled representation learning lose sight of the contextual dependency between the A-O primitive pairs. Inspired by this, we propose a novel A-O disentangled framew…
▽ More
Attribute and object (A-O) disentanglement is a fundamental and critical problem for Compositional Zero-shot Learning (CZSL), whose aim is to recognize novel A-O compositions based on foregone knowledge. Existing methods based on disentangled representation learning lose sight of the contextual dependency between the A-O primitive pairs. Inspired by this, we propose a novel A-O disentangled framework for CZSL, namely Class-specified Cascaded Network (CSCNet). The key insight is to firstly classify one primitive and then specifies the predicted class as a priori for guiding another primitive recognition in a cascaded fashion. To this end, CSCNet constructs Attribute-to-Object and Object-to-Attribute cascaded branches, in addition to a composition branch modeling the two primitives as a whole. Notably, we devise a parametric classifier (ParamCls) to improve the matching between visual and semantic embeddings. By improving the A-O disentanglement, our framework achieves superior results than previous competitive methods.
△ Less
Submitted 13 March, 2024; v1 submitted 9 March, 2024;
originally announced March 2024.
-
Low Complexity Channel Estimation for RIS-Assisted THz Systems with Beam Split
Authors:
Xin Su,
Ruisi He,
Peng Zhang,
Bo Ai
Abstract:
To support extremely high data rates, reconfigurable intelligent surface (RIS)-assisted terahertz (THz) communication is considered to be a promising technology for future sixth-generation networks. However, due to the typical employment of hybrid beamforming architecture in THz systems, as well as the passive nature of RIS which lacks the capability to process pilot signals, obtaining channel sta…
▽ More
To support extremely high data rates, reconfigurable intelligent surface (RIS)-assisted terahertz (THz) communication is considered to be a promising technology for future sixth-generation networks. However, due to the typical employment of hybrid beamforming architecture in THz systems, as well as the passive nature of RIS which lacks the capability to process pilot signals, obtaining channel state information (CSI) is facing significant challenges. To accurately estimate the cascaded channel, we propose a novel low-complexity channel estimation scheme, which includes three steps. Specifically, we first estimate full CSI within a small subset of subcarriers (SCs). Then, we acquire angular information at base station and RIS based on the full CSI. Finally, we derive spatial directions and recover full-CSI for the remaining SCs. Theoretical analysis and simulation results demonstrate that the proposed scheme can achieve superior performance in terms of normalized mean-square-error and exhibit a lower computational complexity compared with the existing algorithms.
△ Less
Submitted 5 March, 2024;
originally announced March 2024.
-
InfiMM-HD: A Leap Forward in High-Resolution Multimodal Understanding
Authors:
Haogeng Liu,
Quanzeng You,
Xiaotian Han,
Yiqi Wang,
Bohan Zhai,
Yongfei Liu,
Yunzhe Tao,
Huaibo Huang,
Ran He,
Hongxia Yang
Abstract:
Multimodal Large Language Models (MLLMs) have experienced significant advancements recently. Nevertheless, challenges persist in the accurate recognition and comprehension of intricate details within high-resolution images. Despite being indispensable for the development of robust MLLMs, this area remains underinvestigated. To tackle this challenge, our work introduces InfiMM-HD, a novel architect…
▽ More
Multimodal Large Language Models (MLLMs) have experienced significant advancements recently. Nevertheless, challenges persist in the accurate recognition and comprehension of intricate details within high-resolution images. Despite being indispensable for the development of robust MLLMs, this area remains underinvestigated. To tackle this challenge, our work introduces InfiMM-HD, a novel architecture specifically designed for processing images of different resolutions with low computational overhead. This innovation facilitates the enlargement of MLLMs to higher-resolution capabilities. InfiMM-HD incorporates a cross-attention module and visual windows to reduce computation costs. By integrating this architectural design with a four-stage training pipeline, our model attains improved visual perception efficiently and cost-effectively. Empirical study underscores the robustness and effectiveness of InfiMM-HD, opening new avenues for exploration in related areas. Codes and models can be found at https://huggingface.co/Infi-MM/infimm-hd
△ Less
Submitted 3 March, 2024;
originally announced March 2024.
-
Infusing Hierarchical Guidance into Prompt Tuning: A Parameter-Efficient Framework for Multi-level Implicit Discourse Relation Recognition
Authors:
Haodong Zhao,
Ruifang He,
Mengnan Xiao,
Jing Xu
Abstract:
Multi-level implicit discourse relation recognition (MIDRR) aims at identifying hierarchical discourse relations among arguments. Previous methods achieve the promotion through fine-tuning PLMs. However, due to the data scarcity and the task gap, the pre-trained feature space cannot be accurately tuned to the task-specific space, which even aggravates the collapse of the vanilla space. Besides, th…
▽ More
Multi-level implicit discourse relation recognition (MIDRR) aims at identifying hierarchical discourse relations among arguments. Previous methods achieve the promotion through fine-tuning PLMs. However, due to the data scarcity and the task gap, the pre-trained feature space cannot be accurately tuned to the task-specific space, which even aggravates the collapse of the vanilla space. Besides, the comprehension of hierarchical semantics for MIDRR makes the conversion much harder. In this paper, we propose a prompt-based Parameter-Efficient Multi-level IDRR (PEMI) framework to solve the above problems. First, we leverage parameter-efficient prompt tuning to drive the inputted arguments to match the pre-trained space and realize the approximation with few parameters. Furthermore, we propose a hierarchical label refining (HLR) method for the prompt verbalizer to deeply integrate hierarchical guidance into the prompt tuning. Finally, our model achieves comparable results on PDTB 2.0 and 3.0 using about 0.1% trainable parameters compared with baselines and the visualization demonstrates the effectiveness of our HLR method.
△ Less
Submitted 22 February, 2024;
originally announced February 2024.
-
Diffusion Model-Based Multiobjective Optimization for Gasoline Blending Scheduling
Authors:
Wenxuan Fang,
Wei Du,
Renchu He,
Yang Tang,
Yaochu Jin,
Gary G. Yen
Abstract:
Gasoline blending scheduling uses resource allocation and operation sequencing to meet a refinery's production requirements. The presence of nonlinearity, integer constraints, and a large number of decision variables adds complexity to this problem, posing challenges for traditional and evolutionary algorithms. This paper introduces a novel multiobjective optimization approach driven by a diffusio…
▽ More
Gasoline blending scheduling uses resource allocation and operation sequencing to meet a refinery's production requirements. The presence of nonlinearity, integer constraints, and a large number of decision variables adds complexity to this problem, posing challenges for traditional and evolutionary algorithms. This paper introduces a novel multiobjective optimization approach driven by a diffusion model (named DMO), which is designed specifically for gasoline blending scheduling. To address integer constraints and generate feasible schedules, the diffusion model creates multiple intermediate distributions between Gaussian noise and the feasible domain. Through iterative processes, the solutions transition from Gaussian noise to feasible schedules while optimizing the objectives using the gradient descent method. DMO achieves simultaneous objective optimization and constraint adherence. Comparative tests are conducted to evaluate DMO's performance across various scales. The experimental results demonstrate that DMO surpasses state-of-the-art multiobjective evolutionary algorithms in terms of efficiency when solving gasoline blending scheduling problems.
△ Less
Submitted 4 February, 2024;
originally announced February 2024.
-
Debiasing Text-to-Image Diffusion Models
Authors:
Ruifei He,
Chuhui Xue,
Haoru Tan,
Wenqing Zhang,
Yingchen Yu,
Song Bai,
Xiaojuan Qi
Abstract:
Learning-based Text-to-Image (TTI) models like Stable Diffusion have revolutionized the way visual content is generated in various domains. However, recent research has shown that nonnegligible social bias exists in current state-of-the-art TTI systems, which raises important concerns. In this work, we target resolving the social bias in TTI diffusion models. We begin by formalizing the problem se…
▽ More
Learning-based Text-to-Image (TTI) models like Stable Diffusion have revolutionized the way visual content is generated in various domains. However, recent research has shown that nonnegligible social bias exists in current state-of-the-art TTI systems, which raises important concerns. In this work, we target resolving the social bias in TTI diffusion models. We begin by formalizing the problem setting and use the text descriptions of bias groups to establish an unsafe direction for guiding the diffusion process. Next, we simplify the problem into a weight optimization problem and attempt a Reinforcement solver, Policy Gradient, which shows sub-optimal performance with slow convergence. Further, to overcome limitations, we propose an iterative distribution alignment (IDA) method. Despite its simplicity, we show that IDA shows efficiency and fast convergence in resolving the social bias in TTI diffusion models. Our code will be released.
△ Less
Submitted 22 February, 2024;
originally announced February 2024.
-
Tables as Texts or Images: Evaluating the Table Reasoning Ability of LLMs and MLLMs
Authors:
Naihao Deng,
Zhenjie Sun,
Ruiqi He,
Aman Sikka,
Yulong Chen,
Lin Ma,
Yue Zhang,
Rada Mihalcea
Abstract:
In this paper, we investigate the effectiveness of various LLMs in interpreting tabular data through different prompting strategies and data formats. Our analyses extend across six benchmarks for table-related tasks such as question-answering and fact-checking. We introduce for the first time the assessment of LLMs' performance on image-based table representations. Specifically, we compare five te…
▽ More
In this paper, we investigate the effectiveness of various LLMs in interpreting tabular data through different prompting strategies and data formats. Our analyses extend across six benchmarks for table-related tasks such as question-answering and fact-checking. We introduce for the first time the assessment of LLMs' performance on image-based table representations. Specifically, we compare five text-based and three image-based table representations, demonstrating the role of representation and prompting on LLM performance. Our study provides insights into the effective use of LLMs on table-related tasks.
△ Less
Submitted 5 June, 2024; v1 submitted 19 February, 2024;
originally announced February 2024.
-
A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation
Authors:
Zhengbo Wang,
Jian Liang,
Lijun Sheng,
Ran He,
Zilei Wang,
Tieniu Tan
Abstract:
Contrastive Language-Image Pretraining (CLIP) has gained popularity for its remarkable zero-shot capacity. Recent research has focused on developing efficient fine-tuning methods, such as prompt learning and adapter, to enhance CLIP's performance in downstream tasks. However, these methods still require additional training time and computational resources, which is undesirable for devices with lim…
▽ More
Contrastive Language-Image Pretraining (CLIP) has gained popularity for its remarkable zero-shot capacity. Recent research has focused on developing efficient fine-tuning methods, such as prompt learning and adapter, to enhance CLIP's performance in downstream tasks. However, these methods still require additional training time and computational resources, which is undesirable for devices with limited resources. In this paper, we revisit a classical algorithm, Gaussian Discriminant Analysis (GDA), and apply it to the downstream classification of CLIP. Typically, GDA assumes that features of each class follow Gaussian distributions with identical covariance. By leveraging Bayes' formula, the classifier can be expressed in terms of the class means and covariance, which can be estimated from the data without the need for training. To integrate knowledge from both visual and textual modalities, we ensemble it with the original zero-shot classifier within CLIP. Extensive results on 17 datasets validate that our method surpasses or achieves comparable results with state-of-the-art methods on few-shot classification, imbalanced learning, and out-of-distribution generalization. In addition, we extend our method to base-to-new generalization and unsupervised learning, once again demonstrating its superiority over competing approaches. Our code is publicly available at \url{https://github.com/mrflogs/ICLR24}.
△ Less
Submitted 6 February, 2024;
originally announced February 2024.
-
Connecting the Dots: Collaborative Fine-tuning for Black-Box Vision-Language Models
Authors:
Zhengbo Wang,
Jian Liang,
Ran He,
Zilei Wang,
Tieniu Tan
Abstract:
With the emergence of pretrained vision-language models (VLMs), considerable efforts have been devoted to fine-tuning them for downstream tasks. Despite the progress made in designing efficient fine-tuning methods, such methods require access to the model's parameters, which can be challenging as model owners often opt to provide their models as a black box to safeguard model ownership. This paper…
▽ More
With the emergence of pretrained vision-language models (VLMs), considerable efforts have been devoted to fine-tuning them for downstream tasks. Despite the progress made in designing efficient fine-tuning methods, such methods require access to the model's parameters, which can be challenging as model owners often opt to provide their models as a black box to safeguard model ownership. This paper proposes a \textbf{C}ollabo\textbf{ra}tive \textbf{F}ine-\textbf{T}uning (\textbf{CraFT}) approach for fine-tuning black-box VLMs to downstream tasks, where one only has access to the input prompts and the output predictions of the model. CraFT comprises two modules, a prompt generation module for learning text prompts and a prediction refinement module for enhancing output predictions in residual style. Additionally, we introduce an auxiliary prediction-consistent loss to promote consistent optimization across these modules. These modules are optimized by a novel collaborative training algorithm. Extensive experiments on few-shot classification over 15 datasets demonstrate the superiority of CraFT. The results show that CraFT achieves a decent gain of about 12\% with 16-shot datasets and only 8,000 queries. Moreover, CraFT trains faster and uses only about 1/80 of the memory footprint for deployment, while sacrificing only 1.62\% compared to the white-box method. Our code is publicly available at https://github.com/mrflogs/CraFT .
△ Less
Submitted 3 June, 2024; v1 submitted 6 February, 2024;
originally announced February 2024.
-
Towards Eliminating Hard Label Constraints in Gradient Inversion Attacks
Authors:
Yanbo Wang,
Jian Liang,
Ran He
Abstract:
Gradient inversion attacks aim to reconstruct local training data from intermediate gradients exposed in the federated learning framework. Despite successful attacks, all previous methods, starting from reconstructing a single data point and then relaxing the single-image limit to batch level, are only tested under hard label constraints. Even for single-image reconstruction, we still lack an anal…
▽ More
Gradient inversion attacks aim to reconstruct local training data from intermediate gradients exposed in the federated learning framework. Despite successful attacks, all previous methods, starting from reconstructing a single data point and then relaxing the single-image limit to batch level, are only tested under hard label constraints. Even for single-image reconstruction, we still lack an analysis-based algorithm to recover augmented soft labels. In this work, we change the focus from enlarging batchsize to investigating the hard label constraints, considering a more realistic circumstance where label smoothing and mixup techniques are used in the training process. In particular, we are the first to initiate a novel algorithm to simultaneously recover the ground-truth augmented label and the input feature of the last fully-connected layer from single-input gradients, and provide a necessary condition for any analytical-based label recovery methods. Extensive experiments testify to the label recovery accuracy, as well as the benefits to the following image reconstruction. We believe soft labels in classification tasks are worth further attention in gradient inversion attacks.
△ Less
Submitted 15 April, 2024; v1 submitted 5 February, 2024;
originally announced February 2024.
-
Knowledge-Assisted Dual-Stage Evolutionary Optimization of Large-Scale Crude Oil Scheduling
Authors:
Wanting Zhang,
Wei Du,
Guo Yu,
Renchu He,
Wenli Du,
Yaochu Jin
Abstract:
With the scaling up of crude oil scheduling in modern refineries, large-scale crude oil scheduling problems (LSCOSPs) emerge with thousands of binary variables and non-linear constraints, which are challenging to be optimized by traditional optimization methods. To solve LSCOSPs, we take the practical crude oil scheduling from a marine-access refinery as an example and start with modeling LSCOSPs…
▽ More
With the scaling up of crude oil scheduling in modern refineries, large-scale crude oil scheduling problems (LSCOSPs) emerge with thousands of binary variables and non-linear constraints, which are challenging to be optimized by traditional optimization methods. To solve LSCOSPs, we take the practical crude oil scheduling from a marine-access refinery as an example and start with modeling LSCOSPs from crude unloading, transportation, crude distillation unit processing, and inventory management of intermediate products. On the basis of the proposed model, a dual-stage evolutionary algorithm driven by heuristic rules (denoted by DSEA/HR) is developed, where the dual-stage search mechanism consists of global search and local refinement. In the global search stage, we devise several heuristic rules based on the empirical operating knowledge to generate a well-performing initial population and accelerate convergence in the mixed variables space. In the local refinement stage, a repair strategy is proposed to move the infeasible solutions towards feasible regions by further optimizing the local continuous variables. During the whole evolutionary process, the proposed dual-stage framework plays a crucial role in balancing exploration and exploitation. Experimental results have shown that DSEA/HR outperforms the state-of-the-art and widely-used mathematical programming methods and metaheuristic algorithms on LSCOSP instances within a reasonable time.
△ Less
Submitted 9 January, 2024;
originally announced January 2024.
-
Can We Trust the Unlabeled Target Data? Towards Backdoor Attack and Defense on Model Adaptation
Authors:
Lijun Sheng,
Jian Liang,
Ran He,
Zilei Wang,
Tieniu Tan
Abstract:
Model adaptation tackles the distribution shift problem with a pre-trained model instead of raw data, becoming a popular paradigm due to its great privacy protection. Existing methods always assume adapting to a clean target domain, overlooking the security risks of unlabeled samples. In this paper, we explore the potential backdoor attacks on model adaptation launched by well-designed poisoning t…
▽ More
Model adaptation tackles the distribution shift problem with a pre-trained model instead of raw data, becoming a popular paradigm due to its great privacy protection. Existing methods always assume adapting to a clean target domain, overlooking the security risks of unlabeled samples. In this paper, we explore the potential backdoor attacks on model adaptation launched by well-designed poisoning target data. Concretely, we provide two backdoor triggers with two poisoning strategies for different prior knowledge owned by attackers. These attacks achieve a high success rate and keep the normal performance on clean samples in the test stage. To defend against backdoor embedding, we propose a plug-and-play method named MixAdapt, combining it with existing adaptation algorithms. Experiments across commonly used benchmarks and adaptation methods demonstrate the effectiveness of MixAdapt. We hope this work will shed light on the safety of learning with unlabeled data.
△ Less
Submitted 11 January, 2024;
originally announced January 2024.
-
Exploring Vacant Classes in Label-Skewed Federated Learning
Authors:
Kuangpu Guo,
Yuhe Ding,
Jian Liang,
Ran He,
Zilei Wang,
Tieniu Tan
Abstract:
Label skews, characterized by disparities in local label distribution across clients, pose a significant challenge in federated learning. As minority classes suffer from worse accuracy due to overfitting on local imbalanced data, prior methods often incorporate class-balanced learning techniques during local training. Although these methods improve the mean accuracy across all classes, we observe…
▽ More
Label skews, characterized by disparities in local label distribution across clients, pose a significant challenge in federated learning. As minority classes suffer from worse accuracy due to overfitting on local imbalanced data, prior methods often incorporate class-balanced learning techniques during local training. Although these methods improve the mean accuracy across all classes, we observe that vacant classes-referring to categories absent from a client's data distribution-remain poorly recognized. Besides, there is still a gap in the accuracy of local models on minority classes compared to the global model. This paper introduces FedVLS, a novel approach to label-skewed federated learning that integrates both vacant-class distillation and logit suppression simultaneously. Specifically, vacant-class distillation leverages knowledge distillation during local training on each client to retain essential information related to vacant classes from the global model. Moreover, logit suppression directly penalizes network logits for non-label classes, effectively addressing misclassifications in minority classes that may be biased toward majority classes. Extensive experiments validate the efficacy of FedVLS, demonstrating superior performance compared to previous state-of-the-art (SOTA) methods across diverse datasets with varying degrees of label skews. Code is available in the supplementary material.
△ Less
Submitted 19 August, 2024; v1 submitted 4 January, 2024;
originally announced January 2024.
-
Low-latency Space-time Supersampling for Real-time Rendering
Authors:
Ruian He,
Shili Zhou,
Yuqi Sun,
Ri Cheng,
Weimin Tan,
Bo Yan
Abstract:
With the rise of real-time rendering and the evolution of display devices, there is a growing demand for post-processing methods that offer high-resolution content in a high frame rate. Existing techniques often suffer from quality and latency issues due to the disjointed treatment of frame supersampling and extrapolation. In this paper, we recognize the shared context and mechanisms between frame…
▽ More
With the rise of real-time rendering and the evolution of display devices, there is a growing demand for post-processing methods that offer high-resolution content in a high frame rate. Existing techniques often suffer from quality and latency issues due to the disjointed treatment of frame supersampling and extrapolation. In this paper, we recognize the shared context and mechanisms between frame supersampling and extrapolation, and present a novel framework, Space-time Supersampling (STSS). By integrating them into a unified framework, STSS can improve the overall quality with lower latency. To implement an efficient architecture, we treat the aliasing and warping holes unified as reshading regions and put forth two key components to compensate the regions, namely Random Reshading Masking (RRM) and Efficient Reshading Module (ERM). Extensive experiments demonstrate that our approach achieves superior visual fidelity compared to state-of-the-art (SOTA) methods. Notably, the performance is achieved within only 4ms, saving up to 75\% of time against the conventional two-stage pipeline that necessitates 17ms.
△ Less
Submitted 17 December, 2023;
originally announced December 2023.
-
How Well Does GPT-4V(ision) Adapt to Distribution Shifts? A Preliminary Investigation
Authors:
Zhongyi Han,
Guanglin Zhou,
Rundong He,
Jindong Wang,
Tailin Wu,
Yilong Yin,
Salman Khan,
Lina Yao,
Tongliang Liu,
Kun Zhang
Abstract:
In machine learning, generalization against distribution shifts -- where deployment conditions diverge from the training scenarios -- is crucial, particularly in fields like climate modeling, biomedicine, and autonomous driving. The emergence of foundation models, distinguished by their extensive pretraining and task versatility, has led to an increased interest in their adaptability to distributi…
▽ More
In machine learning, generalization against distribution shifts -- where deployment conditions diverge from the training scenarios -- is crucial, particularly in fields like climate modeling, biomedicine, and autonomous driving. The emergence of foundation models, distinguished by their extensive pretraining and task versatility, has led to an increased interest in their adaptability to distribution shifts. GPT-4V(ision) acts as the most advanced publicly accessible multimodal foundation model, with extensive applications across various domains, including anomaly detection, video understanding, image generation, and medical diagnosis. However, its robustness against data distributions remains largely underexplored. Addressing this gap, this study rigorously evaluates GPT-4V's adaptability and generalization capabilities in dynamic environments, benchmarking against prominent models like CLIP, LLaVA, and Gemini. We delve into GPT-4V's zero-shot generalization across 13 diverse datasets spanning natural, medical, and molecular domains. We further investigate its adaptability to controlled data perturbations and examine the efficacy of in-context learning as a tool to enhance its adaptation. Our findings delineate GPT-4V's capability boundaries in distribution shifts, shedding light on its strengths and limitations across various scenarios. Importantly, this investigation contributes to our understanding of how AI foundation models generalize to distribution shifts, offering pivotal insights into their adaptability and robustness. The code is publicly available at https://github.com/jameszhou-gl/gpt-4v-distribution-shift.
△ Less
Submitted 25 February, 2024; v1 submitted 12 December, 2023;
originally announced December 2023.
-
Context-Aware Iteration Policy Network for Efficient Optical Flow Estimation
Authors:
Ri Cheng,
Ruian He,
Xuhao Jiang,
Shili Zhou,
Weimin Tan,
Bo Yan
Abstract:
Existing recurrent optical flow estimation networks are computationally expensive since they use a fixed large number of iterations to update the flow field for each sample. An efficient network should skip iterations when the flow improvement is limited. In this paper, we develop a Context-Aware Iteration Policy Network for efficient optical flow estimation, which determines the optimal number of…
▽ More
Existing recurrent optical flow estimation networks are computationally expensive since they use a fixed large number of iterations to update the flow field for each sample. An efficient network should skip iterations when the flow improvement is limited. In this paper, we develop a Context-Aware Iteration Policy Network for efficient optical flow estimation, which determines the optimal number of iterations per sample. The policy network achieves this by learning contextual information to realize whether flow improvement is bottlenecked or minimal. On the one hand, we use iteration embedding and historical hidden cell, which include previous iterations information, to convey how flow has changed from previous iterations. On the other hand, we use the incremental loss to make the policy network implicitly perceive the magnitude of optical flow improvement in the subsequent iteration. Furthermore, the computational complexity in our dynamic network is controllable, allowing us to satisfy various resource preferences with a single trained model. Our policy network can be easily integrated into state-of-the-art optical flow networks. Extensive experiments show that our method maintains performance while reducing FLOPs by about 40%/20% for the Sintel/KITTI datasets.
△ Less
Submitted 5 January, 2024; v1 submitted 12 December, 2023;
originally announced December 2023.
-
Improved Visual Grounding through Self-Consistent Explanations
Authors:
Ruozhen He,
Paola Cascante-Bonilla,
Ziyan Yang,
Alexander C. Berg,
Vicente Ordonez
Abstract:
Vision-and-language models trained to match images with text can be combined with visual explanation methods to point to the locations of specific objects in an image. Our work shows that the localization --"grounding"-- abilities of these models can be further improved by finetuning for self-consistent visual explanations. We propose a strategy for augmenting existing text-image datasets with par…
▽ More
Vision-and-language models trained to match images with text can be combined with visual explanation methods to point to the locations of specific objects in an image. Our work shows that the localization --"grounding"-- abilities of these models can be further improved by finetuning for self-consistent visual explanations. We propose a strategy for augmenting existing text-image datasets with paraphrases using a large language model, and SelfEQ, a weakly-supervised strategy on visual explanation maps for paraphrases that encourages self-consistency. Specifically, for an input textual phrase, we attempt to generate a paraphrase and finetune the model so that the phrase and paraphrase map to the same region in the image. We posit that this both expands the vocabulary that the model is able to handle, and improves the quality of the object locations highlighted by gradient-based visual explanation methods (e.g. GradCAM). We demonstrate that SelfEQ improves performance on Flickr30k, ReferIt, and RefCOCO+ over a strong baseline method and several prior works. Particularly, comparing to other methods that do not use any type of box annotations, we obtain 84.07% on Flickr30k (an absolute improvement of 4.69%), 67.40% on ReferIt (an absolute improvement of 7.68%), and 75.10%, 55.49% on RefCOCO+ test sets A and B respectively (an absolute improvement of 3.74% on average).
△ Less
Submitted 7 December, 2023;
originally announced December 2023.
-
Multimodal Prompt Perceiver: Empower Adaptiveness, Generalizability and Fidelity for All-in-One Image Restoration
Authors:
Yuang Ai,
Huaibo Huang,
Xiaoqiang Zhou,
Jiexiang Wang,
Ran He
Abstract:
Despite substantial progress, all-in-one image restoration (IR) grapples with persistent challenges in handling intricate real-world degradations. This paper introduces MPerceiver: a novel multimodal prompt learning approach that harnesses Stable Diffusion (SD) priors to enhance adaptiveness, generalizability and fidelity for all-in-one image restoration. Specifically, we develop a dual-branch mod…
▽ More
Despite substantial progress, all-in-one image restoration (IR) grapples with persistent challenges in handling intricate real-world degradations. This paper introduces MPerceiver: a novel multimodal prompt learning approach that harnesses Stable Diffusion (SD) priors to enhance adaptiveness, generalizability and fidelity for all-in-one image restoration. Specifically, we develop a dual-branch module to master two types of SD prompts: textual for holistic representation and visual for multiscale detail representation. Both prompts are dynamically adjusted by degradation predictions from the CLIP image encoder, enabling adaptive responses to diverse unknown degradations. Moreover, a plug-in detail refinement module improves restoration fidelity via direct encoder-to-decoder information transformation. To assess our method, MPerceiver is trained on 9 tasks for all-in-one IR and outperforms state-of-the-art task-specific methods across most tasks. Post multitask pre-training, MPerceiver attains a generalized representation in low-level vision, exhibiting remarkable zero-shot and few-shot capabilities in unseen tasks. Extensive experiments on 16 IR tasks underscore the superiority of MPerceiver in terms of adaptiveness, generalizability and fidelity.
△ Less
Submitted 20 March, 2024; v1 submitted 5 December, 2023;
originally announced December 2023.