[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-147701-105":59,"doc-detail-147701-en":131},{"code":4,"msg":5,"data":6},0,"success",[7,13,18,23,28,33,38,43,48,51,55],{"id":8,"doc_module":4,"doc_module_name":9,"category_name":10,"show_sort_weight":11,"slug":12},1,"Document","Story & Novel",90,"story-novel",{"id":14,"doc_module":4,"doc_module_name":9,"category_name":15,"show_sort_weight":16,"slug":17},2,"Literature",80,"literature",{"id":19,"doc_module":4,"doc_module_name":9,"category_name":20,"show_sort_weight":21,"slug":22},4,"Exam",70,"exam",{"id":24,"doc_module":4,"doc_module_name":9,"category_name":25,"show_sort_weight":26,"slug":27},5,"Comic",60,"comic",{"id":29,"doc_module":4,"doc_module_name":9,"category_name":30,"show_sort_weight":31,"slug":32},6,"Technology",50,"technology",{"id":34,"doc_module":4,"doc_module_name":9,"category_name":35,"show_sort_weight":36,"slug":37},7,"Healthcare",40,"healthcare",{"id":39,"doc_module":4,"doc_module_name":9,"category_name":40,"show_sort_weight":41,"slug":42},8,"Research & Report",30,"research-report",{"id":44,"doc_module":4,"doc_module_name":9,"category_name":45,"show_sort_weight":46,"slug":47},9,"Religion & Spirituality",20,"religion-spirituality",{"id":46,"doc_module":4,"doc_module_name":9,"category_name":49,"show_sort_weight":46,"slug":50},"World Cup","world-cup",{"id":52,"doc_module":4,"doc_module_name":9,"category_name":53,"show_sort_weight":52,"slug":54},10,"Lifestyle","lifestyle",{"id":56,"doc_module":4,"doc_module_name":9,"category_name":57,"show_sort_weight":24,"slug":58},19,"General","general",{"code":4,"msg":60,"data":61},"ok",{"site_id":62,"language":63,"slug":64,"title":65,"keywords":66,"description":67,"schema_data":68,"social_meta":124,"head_meta":126,"extra_data":128,"updated_unix":130},105,"en","learning-probabilistic-models-for-visual-motion-doctor-of-philosophy-thesis","Learning Probabilistic Models for Visual Motion - Doctor of Philosophy Thesis","","A fundamental goal of computer vision is analyzing motion, from tracking rigid objects in images to recovering full pose and articulated behavior of nonrigid interacting objects. This thesis addresses the limitation of narrowly engineered systems by proposing machine-learning solutions that adapt automatically to new motion inputs. It develops two tracking methods for learning appearance and dynamics from limited or labeled data, and a probabilistic graphical model for discovering non-rigid structure from feature trajectories.",{"@graph":69,"@context":123},[70,84,106],{"@type":71,"itemListElement":72},"BreadcrumbList",[73,77,79,82],{"item":74,"name":75,"@type":76,"position":8},"https://docshare.wps.com","Home","ListItem",{"item":78,"name":9,"@type":76,"position":14},"https://docshare.wps.com/document/",{"item":80,"name":40,"@type":76,"position":81},"https://docshare.wps.com/document/research-report/",3,{"item":83,"name":65,"@type":76,"position":19},"https://docshare.wps.com/document/learning-probabilistic-models-for-visual-motion-doctor-of-philosophy-thesis/147701/",{"url":83,"name":65,"@type":85,"image":86,"author":91,"headline":65,"publisher":94,"fileFormat":97,"inLanguage":63,"description":67,"dateModified":98,"datePublished":99,"encodingFormat":97,"isAccessibleForFree":100,"interactionStatistic":101},"DigitalDocument",{"url":87,"@type":88,"width":89,"height":90},"https://docshare.wps.com/thumbnails/learning-probabilistic-models-for-visual-motion-doctor-of-philosophy-thesis/147701.png","ImageObject",300,407,{"name":92,"@type":93},"Lucas Martin","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-05","2026-08-26",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",16,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What central problem does the thesis address in computer vision?","Question",{"text":113,"@type":114},"The thesis targets motion analysis, extending capabilities beyond rigid object tracking toward learning-based recovery of pose and nonrigid articulated structure from video or trajectory data.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does the work approach learning for rigid object tracking?",{"text":118,"@type":114},"It presents two complementary approaches: learning an appearance model from very limited training data (using PCA) and a data-driven method that learns appropriate appearance and dynamics models from labeled video sequences.",{"name":120,"@type":111,"acceptedAnswer":121},"How are non-rigid structures and poses recovered from feature trajectories?",{"text":122,"@type":114},"The thesis models observed motion using probabilistic graphical structures (“probabilistic stick figures”), assuming stick lengths and connectivity are fixed while relative joint angles can change over time.","https://schema.org",{"og:url":83,"og:type":125,"og:title":65,"og:site_name":95,"og:description":67},"article",{"robots":127,"canonical":83},"index,follow",{"doc_id":129,"site_id":62},147701,1787768532,{"code":4,"msg":5,"data":132},{"doc_id":129,"user_id":133,"nickname":92,"user_avatar":134,"doc_module":4,"category_id":39,"category_name":40,"doc_title":65,"doc_description":67,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":105,"is_deleted":4,"is_public":8,"is_downloadable":8,"audit_status":8,"page_count":140,"language":141,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":67,"update_tm":130,"read_time":145,"preview_page_count":4},8796095360427,"https://ap-avatar.wpscdn.com/davatar_994ba38a5ba835b3df7d355c54d3ed8d","Learning Probabilistic Models for Visual Motion  \nby  \nDavid Alexander Ross  \nA thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Graduate Department of Computer Science University of Toronto  \nCopyright 􀀍c 2008 by David Alexander Ross  \nAbstract  \nLearning Probabilistic Models for Visual Motion  \nDavid Alexander Ross  \nDoctor of Philosophy  \nGraduate Department of Computer Science  \nUniversity of Toronto  \n2008  \nA fundamental goal of computer vision is the ability to analyze motion. This can range from the simple task of locating or tracking a single rigid object as it moves across an image plane, to recovering the full pose parameters of a collection of nonrigid objects interacting in a scene. The current state of computer vision research, as with the preponderance of challenges that comprise \\arti􀀌cial intelligence\", is that the abilities of humans can only be matched in very narrow domains by carefully and speci􀀌cally engineered systems.  \nThe key to broadening the applicability of these successful systems is to imbue them with the 􀀍exibility to handle new inputs, and to adapt automatically without the manual intervention of human engineers. In this research we attempt to address this challenge by proposing solutions to motion analysis tasks that are based on machine learning.  \nWe begin by addressing the challenge of tracking a rigid object in video, presenting two complementary approaches. First we explore the problem of learning a particular choice of appearance model|principal components analysis (PCA)|from a very limited set of training data. However, PCA is far from the only appearance model available. This raises the question: given a new tracking task, how should one select the most-appropriate models of appearance and dynamics? Our second approach proposes a data-driven solution to this problem, allowing the choice of models, along with their parameters, to be learned from a labelled video sequence.  \nNext we consider motion analysis at a higher-level of organization. Given a set of trajectories obtained by tracking various feature points, how can we discover the underlying non-rigid structure of the object or objects? We propose a solution that models the observed sequence in terms of probabilistic \\stick 􀀌gures\", under the assumption that the relative joint angles between sticks can change over time, but their lengths and connectivities are 􀀌xed. We demonstrate the ability to recover the invariant structure and the pose of articulated objects from a number of challenging datasets.  \nAcknowledgements  \nThis thesis could not have been completed without the invaluable contributions of my advisor Richard Zemel, and my collaborators Jongwoo Lim, Ruei-Sung Lin, Edward Meeds, Simon Osindero, Sam Roweis, Daniel Tarlow, and Ming-Hsuan Yang.  \nAdditional assistance with experiments and 􀀌gures was provided by Thomas ElMaraghi, Bohyung Han, Allan Jepson, Volodymyr Mnih, Fabian Wauthier, and Karolina B􀁿uchner.  \nThe motion capture data used in this project was provided by the Biomotion Lab, Queen's University, Canada, and the Carnegie Mellon University Motion Capture Database [http://mocap.cs.cmu.edu/](http://mocap.cs.cmu.edu/) (created with funding from NSF EIA-0196217).  \nWork presented in this thesis was supported in part by the Government of Canada, through a Canada Graduate Scholarship, and by an internship at Honda Research Institute, USA.  \nContents  \n1 Introduction 1  \n1.1 Visual Tracking ................................ 2  \n1.1.1 Incremental Learning of Appearance ................ 3  \n1.1.2 Tracking with Learned Combinations of Discriminative Features . 4  \n1.2 Learning Articulated Structure and Motion ................. 5  \n1.3 Organization ................................. 6  \n2 Incremental Learning for Robust Visual Tracking 8  \n2.1 Introduction .................................. 9  \n2.2 Related Work and Motivation ........................ 11  \n2.3 Incremental Learning fo","cbCaia5ZyQlzRtuw","https://ap.wps.com/l/cbCaia5ZyQlzRtuw","pdf",9206613,168,"English","# Introduction\n## Visual Tracking\n## Learning Articulated Structure and Motion\n## Organization\n# Incremental Learning for Robust Visual Tracking\n## Introduction\n## Related Work and Motivation\n## Incremental Learning for Tracking\n## Implementation and Experiments\n## Conclusions and Future Work\n# Combining Discriminative Features\n## Introduction\n## Model\n## Inference\n## Learning\n## Related Work\n## A visual tracking application\n## Discussion\n# Learning Articulated Structure and Motion\n## Introduction\n## Related Work\n## Model\n## Learning\n## Experimental Results and Analysis","[{\"question\":\"What central problem does the thesis address in computer vision?\",\"answer\":\"The thesis targets motion analysis, extending capabilities beyond rigid object tracking toward learning-based recovery of pose and nonrigid articulated structure from video or trajectory data.\"},{\"question\":\"How does the work approach learning for rigid object tracking?\",\"answer\":\"It presents two complementary approaches: learning an appearance model from very limited training data (using PCA) and a data-driven method that learns appropriate appearance and dynamics models from labeled video sequences.\"},{\"question\":\"How are non-rigid structures and poses recovered from feature trajectories?\",\"answer\":\"The thesis models observed motion using probabilistic graphical structures (“probabilistic stick figures”), assuming stick lengths and connectivity are fixed while relative joint angles can change over time.\"}]","Learning Probabilistic Models for Visual Motion - Doctor of Philosophy Thesis | PDF",423]