{"id":186897,"date":"2011-09-28T00:00:00","date_gmt":"2011-10-17T12:06:13","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/structured-prediction-models-for-high-level-computer-vision-tasks\/"},"modified":"2016-08-22T11:30:37","modified_gmt":"2016-08-22T18:30:37","slug":"structured-prediction-models-for-high-level-computer-vision-tasks","status":"publish","type":"msr-video","link":"https:\/\/www.microsoft.com\/en-us\/research\/video\/structured-prediction-models-for-high-level-computer-vision-tasks\/","title":{"rendered":"Structured Prediction Models for High-level Computer Vision Tasks"},"content":{"rendered":"<div class=\"asset-content\">\n<p>Rich statistical models have revolutionized computer vision research:<\/p>\n<p>graphical models and structured prediction in particular are now commonly used<br \/>\ntools to address hard computer vision problems.  I discuss what distinguishes<br \/>\nthese computer vision problems from other machine learning problems and how<br \/>\nthis poses unique challenges.<\/p>\n<p>I argue that most computer vision models are misspecified and discuss the<br \/>\nconsequences of popular estimators in this case, concluding that we either<br \/>\nhave to use non-parametric models or use estimators robust to<br \/>\nmisspecification.<\/p>\n<p>As one possible solution, I propose a novel discrete random field model<br \/>\napplicable to a large number of computer vision tasks.  The model is<br \/>\nconditionally specified, non-parametric, and able to represent complex label<br \/>\ninteractions, yet it can be trained from hundreds of images in minutes on a<br \/>\nsingle machine.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>Rich statistical models have revolutionized computer vision research: graphical models and structured prediction in particular are now commonly used tools to address hard computer vision problems. I discuss what distinguishes these computer vision problems from other machine learning problems and how this poses unique challenges. I argue that most computer vision models are misspecified and [&hellip;]<\/p>\n","protected":false},"featured_media":196433,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","msr_hide_image_in_river":0,"footnotes":""},"research-area":[],"msr-video-type":[],"msr-locale":[268875],"msr-post-option":[],"msr-session-type":[],"msr-impact-theme":[],"msr-pillar":[],"msr-episode":[],"msr-research-theme":[],"class_list":["post-186897","msr-video","type-msr-video","status-publish","has-post-thumbnail","hentry","msr-locale-en_us"],"msr_download_urls":"","msr_external_url":"https:\/\/youtu.be\/AlRBDnRiXNo","msr_secondary_video_url":"","msr_video_file":"","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/186897","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-video"}],"version-history":[{"count":0,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video\/186897\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media\/196433"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=186897"}],"wp:term":[{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=186897"},{"taxonomy":"msr-video-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-video-type?post=186897"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=186897"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=186897"},{"taxonomy":"msr-session-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-session-type?post=186897"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=186897"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=186897"},{"taxonomy":"msr-episode","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-episode?post=186897"},{"taxonomy":"msr-research-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-theme?post=186897"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}