Content deleted Content added
Citation bot (talk | contribs) Add: date. | Use this bot. Report bugs. | Suggested by Abductive | #UCB_webform 2109/3850 |
HirowoWiki (talk | contribs) m En dash Tags: Visual edit Mobile edit Mobile web edit |
||
| (22 intermediate revisions by 20 users not shown) | |||
Line 1:
{{Short description|Image dataset}}
{{Use dmy dates|date=September 2019}}The '''ImageNet''' project is a large visual [[database]] designed for use in [[Outline of object recognition|visual object recognition software]] research. More than 14 million<ref name="New Scientist">{{cite news|title=New computer vision challenge wants to teach robots to see in 3D|url=https://www.newscientist.com/article/2127131-new-computer-vision-challenge-wants-to-teach-robots-to-see-in-3d/|access-date=3 February 2018|work=New Scientist|date=7 April 2017}}</ref><ref name="nytimes 2012">{{cite news|last1=Markoff|first1=John|title=For Web Images, Creating New Technology to Seek and Find|url=https://www.nytimes.com/2012/11/20/science/for-web-images-creating-new-technology-to-seek-and-find.html|access-date=3 February 2018|work=The New York Times|date=19 November 2012}}</ref> images have been hand-annotated by the project to indicate what objects are pictured and in at least one million of the images, bounding boxes are also provided.<ref>{{
▲The '''ImageNet''' project is a large visual [[database]] designed for use in [[Outline of object recognition|visual object recognition software]] research. More than 14 million<ref name="New Scientist">{{cite news|title=New computer vision challenge wants to teach robots to see in 3D|url=https://www.newscientist.com/article/2127131-new-computer-vision-challenge-wants-to-teach-robots-to-see-in-3d/|access-date=3 February 2018|work=New Scientist|date=7 April 2017}}</ref><ref name="nytimes 2012">{{cite news|last1=Markoff|first1=John|title=For Web Images, Creating New Technology to Seek and Find|url=https://www.nytimes.com/2012/11/20/science/for-web-images-creating-new-technology-to-seek-and-find.html|access-date=3 February 2018|work=The New York Times|date=19 November 2012}}</ref> images have been hand-annotated by the project to indicate what objects are pictured and in at least one million of the images, bounding boxes are also provided.<ref>{{cite web|title=ImageNet Summary and Statistics|url=http://image-net.org/about-stats|publisher=ImageNet|access-date=22 June 2016}}</ref> ImageNet contains more than 20,000 categories,<ref name="nytimes 2012"/> with a typical category, such as "balloon" or "strawberry", consisting of several hundred images.<ref name=economist>{{cite news|title=From not working to neural networking|url=https://www.economist.com/news/special-report/21700756-artificial-intelligence-boom-based-old-idea-modern-twist-not|access-date=3 February 2018|newspaper=The Economist|date=25 June 2016}}</ref> The database of annotations of third-party image [[URL]]s is freely available directly from ImageNet, though the actual images are not owned by ImageNet.<ref>{{cite web|title=ImageNet Overview|url=http://image-net.org/about-overview|publisher=ImageNet|access-date=22 June 2016}}</ref> Since 2010, the ImageNet project runs an annual software contest, the ImageNet Large Scale Visual Recognition Challenge ([[#History_of_the_ImageNet_challenge|ILSVRC]]), where software programs compete to correctly classify and detect objects and scenes. The challenge uses a "trimmed" list of one thousand non-overlapping classes.<ref name=ILJVRC-2015/>
==Significance for deep learning==
On 30 September 2012, a [[convolutional neural network]] (CNN) called [[AlexNet]]<ref name=":0">{{Cite journal|last1=Krizhevsky|first1=Alex|last2=Sutskever|first2=Ilya|last3=Hinton|first3=Geoffrey E.|access-date=24 May 2017|title=ImageNet classification with deep convolutional neural networks|url=https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf|journal=Communications of the ACM|volume=60|issue=6|date=June 2017|pages=84–90|doi=10.1145/3065386|s2cid=195908774|issn=0001-0782|doi-access=free}}</ref> achieved a top-5 error of 15.3% in the ImageNet 2012 Challenge, more than 10.8 percentage points lower than that of the runner up.
In 2015, AlexNet was outperformed by [[Microsoft]]'s [[ResNets|very deep CNN]] with over 100 layers, which won the ImageNet 2015 contest.<ref name="microsoft2015">{{cite
==History of the database==
AI researcher [[Fei-Fei Li]] began working on the idea for ImageNet in 2006. At a time when most AI research focused on models and algorithms, Li wanted to expand and improve the data available to train AI algorithms.<ref name="WiredQuest">{{Cite magazine |url=https://www.wired.com/story/fei-fei-li-artificial-intelligence-humanity/ |title=Fei-Fei Li's Quest to Make AI Better for Humanity |last=Hempel |first=Jesse |magazine=Wired |quote=When Li, who had moved back to Princeton to take a job as an assistant professor in 2007, talked up her idea for ImageNet, she had a hard time getting faculty members to help out. Finally, a professor who specialized in computer architecture agreed to join her as a collaborator. |date=13 November 2018 |access-date=5 May 2019}}</ref> In 2007, Li met with Princeton professor [[Christiane Fellbaum]], one of the creators of [[WordNet]], to discuss the project. As a result of this meeting, Li went on to build ImageNet starting from the word database of WordNet and using many of its features.<ref name="Gershgorn"/>
As an assistant professor at [[Princeton University|Princeton]], Li assembled a team of researchers to work on the ImageNet project. They used [[Amazon Mechanical Turk]] to help with the classification of images.<ref name="Gershgorn"/>
They presented their database for the first time as a poster at the 2009 [[Conference on Computer Vision and Pattern Recognition]] (CVPR) in Florida.<ref name="Gershgorn">{{cite web |url=https://qz.com/1034972/the-data-that-changed-the-direction-of-ai-research-and-possibly-the-world/ |title=The data that transformed AI research—and possibly the world |last=Gershgorn |first=Dave |date=26 July 2017 |website=Quartz |publisher=Atlantic Media Co.|quote=Having read about WordNet's approach, Li met with professor Christiane Fellbaum, a researcher influential in the continued work on WordNet, during a 2006 visit to Princeton. |access-date=26 July 2017 }}</ref><ref>{{Citation |last1=Deng |first1=Jia |last2=Dong |first2=Wei |last3=Socher |first3=Richard |last4=Li |first4=Li-Jia |last5=Li |first5=Kai |last6=Fei-Fei |first6=Li |contribution=ImageNet: A Large-Scale Hierarchical Image Database |year=2009 |title=2009 conference on Computer Vision and Pattern Recognition |contribution-url=http://www.image-net.org/papers/imagenet_cvpr09.pdf |access-date=26 July 2017 |archive-date=15 January 2021 |archive-url=https://web.archive.org/web/20210115185228/http://www.image-net.org/papers/imagenet_cvpr09.pdf |url-status=dead }}</ref><ref>{{Citation|last=Li|first=Fei-Fei|title=How we're teaching computers to understand pictures|date=23 March 2015 |url=https://www.ted.com/talks/fei_fei_li_how_we_re_teaching_computers_to_understand_pictures?language=en|access-date=16 December 2018}}</ref>
==Dataset==
ImageNet [[crowdsources]] its annotation process. Image-level annotations indicate the presence or absence of an object class in an image, such as "there are tigers in this image" or "there are no tigers in this image". Object-level annotations provide a bounding box around the (visible part of the) indicated object. ImageNet uses a variant of the broad [[WordNet]] schema to categorize objects, augmented with 120 categories of [[dog breeds]] to showcase fine-grained classification.<ref name=ILJVRC-2015>Olga Russakovsky*, Jia Deng*, Hao Su, Jonathan Krause, Sanjeev Satheesh, Sean Ma, Zhiheng Huang, [[Andrej Karpathy]], Aditya Khosla, Michael Bernstein, Alexander C. Berg and Li Fei-Fei. (* = equal contribution) ImageNet Large Scale Visual Recognition Challenge. IJCV, 2015.</ref> One downside of WordNet use is the categories may be more "elevated" than would be optimal for ImageNet: "Most people are more interested in Lady Gaga or the iPod Mini than in this rare kind of [[diplodocus]]."{{Clarify|date=August 2019}}<!-- elevated? --> In 2012, ImageNet was the world's largest academic user of [[Amazon Mechanical Turk|Mechanical Turk]]. The average worker identified 50 images per minute.<ref name="nytimes 2012"/>
== Subsets of the dataset ==
There are various subsets of the ImageNet dataset used in various context. One of the most highly used subset of ImageNet is the "ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2012–2017 image classification and localization dataset". This is also referred to in the research literature as ImageNet-1K or ILSVRC2017, reflecting the original ILSVRC challenge that involved 1,000 classes. ImageNet-1K contains 1,281,167 training images, 50,000 validation images and 100,000 test images.<ref>{{Cite web |title=ImageNet |url=https://www.image-net.org/download.php |access-date=2022-10-19 |website=www.image-net.org}}</ref> The full original dataset is referred to as ImageNet-21K. ImageNet-21k contains 14,197,122 images divided into 21,841 classes. Some papers round this up and name it ImageNet-22k.<ref>{{cite arXiv |last1=Ridnik |first1=Tal |last2=Ben-Baruch |first2=Emanuel |last3=Noy |first3=Asaf |last4=Zelnik-Manor |first4=Lihi |date=2021-08-05 |title=ImageNet-21K Pretraining for the Masses |class=cs.CV |eprint=2104.10972 }}</ref>
==History of the ImageNet challenge==
[[File:ImageNet_error_rate_history_(just_systems).svg|thumb|Error rate history on ImageNet (showing best result per team and up to 10 entries per year)]]
The ILSVRC aims to "follow in the footsteps" of the smaller-scale PASCAL VOC challenge, established in 2005, which contained only about 20,000 images and twenty object classes.<ref name="ILJVRC-2015" /> To "democratize" ImageNet, Fei-Fei Li proposed to the PASCAL VOC team a collaboration, beginning in 2010, where research teams would evaluate their algorithms on the given data set, and compete to achieve higher accuracy on several visual recognition tasks.<ref name="Gershgorn"/>
The resulting annual competition is now known as the ImageNet Large Scale Visual Recognition Challenge (ILSVRC). The ILSVRC uses a "trimmed" list of only 1000 image categories or "classes", including 90 of the 120 dog breeds classified by the full ImageNet schema.<ref name="ILJVRC-2015" /> The 2010s saw dramatic progress in image processing. Around 2011, a good ILSVRC classification top-5 error rate was 25%. In 2012, a deep [[Convolutional neural network|convolutional neural net]] called [[AlexNet]] achieved 16%; in the next couple of years, top-5 error rates fell to a few percent.<ref>{{cite news|last1=Robbins|first1=Martin|title=Does an AI need to make love to Rembrandt's girlfriend to make art?|url=https://www.theguardian.com/science/2016/may/06/does-an-ai-need-to-make-love-to-rembrandts-girlfriend-to-make-art|access-date=22 June 2016|work=The Guardian|date=6 May 2016}}</ref> While the 2012 breakthrough "combined pieces that were all there before", the dramatic quantitative improvement marked the start of an industry-wide artificial intelligence boom.<ref name="economist" /> By 2015, researchers at Microsoft reported that their CNNs exceeded human ability at the narrow ILSVRC tasks.<ref name="microsoft2015" /><ref>{{cite news|last1=Markoff|first1=John|title=A Learning Advance in Artificial Intelligence Rivals Human Abilities|url=https://www.nytimes.com/2015/12/11/science/an-advance-in-artificial-intelligence-rivals-human-vision-abilities.html|access-date=22 June 2016|work=The New York Times|date=10 December 2015}}</ref> However, as one of the challenge's organizers, [[Olga Russakovsky]], pointed out in 2015, the programs only have to identify images as belonging to one of a thousand categories; humans can recognize a larger number of categories, and also (unlike the programs) can judge the context of an image.<ref>{{cite news|last1=Aron|first1=Jacob|title=Forget the Turing test – there are better ways of judging AI|url=https://www.newscientist.com/article/dn28206-forget-the-turing-test-there-are-better-ways-of-judging-ai/|access-date=22 June 2016|work=New Scientist|date=21 September 2015}}</ref>
By 2014, more than fifty institutions participated in the ILSVRC.<ref name=ILJVRC-2015 /> In 2017, 29 of 38 competing teams had greater than 95% accuracy.<ref>{{cite news|last1=Gershgorn|first1=Dave|title=The Quartz guide to artificial intelligence: What is it, why is it important, and should we be afraid?|url=https://qz.com/1046350/the-quartz-guide-to-artificial-intelligence-what-is-it-why-is-it-important-and-should-we-be-afraid/|access-date=3 February 2018|work=Quartz|date=10 September 2017}}</ref> In 2017 ImageNet stated it would roll out a new, much more difficult
== Bias in ImageNet ==
A study of the history of the multiple layers ([[Taxonomy (general)|taxonomy]], object classes and labeling) of ImageNet and WordNet in 2019 described how [[Algorithmic bias|bias]]{{clarification needed|date=December 2023}} is deeply embedded in most classification approaches for
== See also ==
| |||