-
Multi Player Tracking in Ice Hockey with Homographic Projections
Authors:
Harish Prakash,
Jia Cheng Shang,
Ken M. Nsiempba,
Yuhao Chen,
David A. Clausi,
John S. Zelek
Abstract:
Multi Object Tracking (MOT) in ice hockey pursues the combined task of localizing and associating players across a given sequence to maintain their identities. Tracking players from monocular broadcast feeds is an important computer vision problem offering various downstream analytics and enhanced viewership experience. However, existing trackers encounter significant difficulties in dealing with…
▽ More
Multi Object Tracking (MOT) in ice hockey pursues the combined task of localizing and associating players across a given sequence to maintain their identities. Tracking players from monocular broadcast feeds is an important computer vision problem offering various downstream analytics and enhanced viewership experience. However, existing trackers encounter significant difficulties in dealing with occlusions, blurs, and agile player movements prevalent in telecast feeds. In this work, we propose a novel tracking approach by formulating MOT as a bipartite graph matching problem infused with homography. We disentangle the positional representations of occluded and overlapping players in broadcast view, by mapping their foot keypoints to an overhead rink template, and encode these projected positions into the graph network. This ensures reliable spatial context for consistent player tracking and unfragmented tracklet prediction. Our results show considerable improvements in both the IDsw and IDF1 metrics on the two available broadcast ice hockey datasets.
△ Less
Submitted 22 May, 2024;
originally announced May 2024.
-
Distribution and Depth-Aware Transformers for 3D Human Mesh Recovery
Authors:
Jerrin Bright,
Bavesh Balaji,
Harish Prakash,
Yuhao Chen,
David A Clausi,
John Zelek
Abstract:
Precise Human Mesh Recovery (HMR) with in-the-wild data is a formidable challenge and is often hindered by depth ambiguities and reduced precision. Existing works resort to either pose priors or multi-modal data such as multi-view or point cloud information, though their methods often overlook the valuable scene-depth information inherently present in a single image. Moreover, achieving robust HMR…
▽ More
Precise Human Mesh Recovery (HMR) with in-the-wild data is a formidable challenge and is often hindered by depth ambiguities and reduced precision. Existing works resort to either pose priors or multi-modal data such as multi-view or point cloud information, though their methods often overlook the valuable scene-depth information inherently present in a single image. Moreover, achieving robust HMR for out-of-distribution (OOD) data is exceedingly challenging due to inherent variations in pose, shape and depth. Consequently, understanding the underlying distribution becomes a vital subproblem in modeling human forms. Motivated by the need for unambiguous and robust human modeling, we introduce Distribution and depth-aware human mesh recovery (D2A-HMR), an end-to-end transformer architecture meticulously designed to minimize the disparity between distributions and incorporate scene-depth leveraging prior depth information. Our approach demonstrates superior performance in handling OOD data in certain scenarios while consistently achieving competitive results against state-of-the-art HMR methods on controlled datasets.
△ Less
Submitted 13 March, 2024;
originally announced March 2024.
-
Jersey Number Recognition using Keyframe Identification from Low-Resolution Broadcast Videos
Authors:
Bavesh Balaji,
Jerrin Bright,
Harish Prakash,
Yuhao Chen,
David A Clausi,
John Zelek
Abstract:
Player identification is a crucial component in vision-driven soccer analytics, enabling various downstream tasks such as player assessment, in-game analysis, and broadcast production. However, automatically detecting jersey numbers from player tracklets in videos presents challenges due to motion blur, low resolution, distortions, and occlusions. Existing methods, utilizing Spatial Transformer Ne…
▽ More
Player identification is a crucial component in vision-driven soccer analytics, enabling various downstream tasks such as player assessment, in-game analysis, and broadcast production. However, automatically detecting jersey numbers from player tracklets in videos presents challenges due to motion blur, low resolution, distortions, and occlusions. Existing methods, utilizing Spatial Transformer Networks, CNNs, and Vision Transformers, have shown success in image data but struggle with real-world video data, where jersey numbers are not visible in most of the frames. Hence, identifying frames that contain the jersey number is a key sub-problem to tackle. To address these issues, we propose a robust keyframe identification module that extracts frames containing essential high-level information about the jersey number. A spatio-temporal network is then employed to model spatial and temporal context and predict the probabilities of jersey numbers in the video. Additionally, we adopt a multi-task loss function to predict the probability distribution of each digit separately. Extensive evaluations on the SoccerNet dataset demonstrate that incorporating our proposed keyframe identification module results in a significant 37.81% and 37.70% increase in the accuracies of 2 different test sets with domain gaps. These results highlight the effectiveness and importance of our approach in tackling the challenges of automatic jersey number detection in sports videos.
△ Less
Submitted 12 September, 2023;
originally announced September 2023.
-
Advaita: Bug Duplicity Detection System
Authors:
Amit Kumar,
Manohar Madanu,
Hari Prakash,
Lalitha Jonnavithula,
Srinivasa Rao Aravilli
Abstract:
Bugs are prevalent in software development. To improve software quality, bugs are filed using a bug tracking system. Properties of a reported bug would consist of a headline, description, project, product, component that is affected by the bug and the severity of the bug. Duplicate bugs rate (% of duplicate bugs) are in the range from single digit (1 to 9%) to double digits (40%) based on the prod…
▽ More
Bugs are prevalent in software development. To improve software quality, bugs are filed using a bug tracking system. Properties of a reported bug would consist of a headline, description, project, product, component that is affected by the bug and the severity of the bug. Duplicate bugs rate (% of duplicate bugs) are in the range from single digit (1 to 9%) to double digits (40%) based on the product maturity , size of the code and number of engineers working on the project. Duplicate bugs range are between 9% to 39% in some of the open source projects like Eclipse, Firefox etc. Detection of duplicity deals with identifying whether any two bugs convey the same meaning. This detection of duplicates helps in de-duplication. Detecting duplicate bugs help reduce triaging efforts and saves time for developers in fixing the issues. Traditional natural language processing techniques are less accurate in identifying similarity between sentences. Using the bug data present in a bug tracking system, various approaches were explored including several machine learning algorithms, to obtain a viable approach that can identify duplicate bugs, given a pair of sentences(i.e. the respective bug descriptions). This approach considers multiple sets of features viz. basic text statistical features, semantic features and contextual features. These features are extracted from the headline, description and component and are subsequently used to train a classification algorithm.
△ Less
Submitted 23 January, 2020;
originally announced January 2020.
-
Standard Quantum Teleportation of an Arbitrary N-Qubit State, Non-Existence of Magic Basis and Existence of Magic Partial Bases for 2N Entangled Qubit States with N>1
Authors:
Hari Prakash,
Vikram Verma
Abstract:
We present a simple and precise protocol for standard quantum teleportation of N-qubit state, considering the most general resource q-channel and Bell states. We find condition on these states for perfect teleportation and give explicitly the unitary transformation required to be done by Bob for achieving perfect teleportation. We discuss connection of our simple theory with the complicated relate…
▽ More
We present a simple and precise protocol for standard quantum teleportation of N-qubit state, considering the most general resource q-channel and Bell states. We find condition on these states for perfect teleportation and give explicitly the unitary transformation required to be done by Bob for achieving perfect teleportation. We discuss connection of our simple theory with the complicated related work on this subject and with character matrix, transformation, judgment and kernel operators defined in this context. We also prove that the magic basis discussed by Hill and Wootters [Phys. Rev. Lett. 78 (1997) 5022] does not exist for entangled 2N-qubit states with N > 1 but magic partial bases, similar to those discussed recently by Prakash and Maurya [Optics Commun. 284 (2011) 5024] do exist. We give explicitly all magic partial bases for N = 2.
△ Less
Submitted 6 October, 2011;
originally announced October 2011.
