{"id":168307,"date":"2015-06-01T00:00:00","date_gmt":"2015-06-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/learning-an-efficient-model-of-hand-shape-variation-from-depth-images\/"},"modified":"2018-10-16T20:14:17","modified_gmt":"2018-10-17T03:14:17","slug":"learning-an-efficient-model-of-hand-shape-variation-from-depth-images","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/learning-an-efficient-model-of-hand-shape-variation-from-depth-images\/","title":{"rendered":"Learning an Efficient Model of Hand Shape Variation from Depth Images"},"content":{"rendered":"<div class=\"asset-content\">\n<p>We describe how to learn a compact and efficient model of the surface deformation of human hands. The model is built from a set of noisy depth images of a diverse set of subjects performing different poses with their hands. We represent the observed surface using Loop subdivision of a control mesh that is deformed by our learned parametric shape and pose model. The model simultaneously accounts for variation in subject-specific shape and subject-agnostic pose. Specifically, hand shape is parameterized as a linear combination of a mean mesh in a neutral pose with a small number of offset vectors. This mesh is then articulated using standard linear blend skinning (LBS) to generate the control mesh of a subdivision surface. We define an energy that encourages each depth pixel to be explained by our model, and the use of a smooth subdivision surface allows us to optimize for all parameters jointly from a rough initialization. The efficacy of our method is demonstrated using both synthetic and real data, where it is shown that hand shape variation can be represented using only a small number of basis components. We compare with other approaches including PCA and show a substantial improvement in the representational power of our model, while maintaining the efficiency of a linear shape basis.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>We describe how to learn a compact and efficient model of the surface deformation of human hands. The model is built from a set of noisy depth images of a diverse set of subjects performing different poses with their hands. We represent the observed surface using Loop subdivision of a control mesh that is deformed [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13556,13562,13551,13554],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-168307","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-artificial-intelligence","msr-research-area-computer-vision","msr-research-area-graphics-and-multimedia","msr-research-area-human-computer-interaction","msr-locale-en_us"],"msr_publishername":"IEEE - Institute of Electrical and Electronics Engineers","msr_edition":"CVPR","msr_affiliation":"","msr_published_date":"2015-06-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"204312","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"Video.mp4","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Video.mp4","id":204312,"label_id":0},{"type":"file","title":"HandShapeVariation.pdf","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/HandShapeVariation.pdf","id":204311,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":204312,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/Video.mp4"},{"id":204311,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2016\/02\/HandShapeVariation.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"samehk","user_id":33504,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=samehk"},{"type":"user_nicename","value":"jota","user_id":32403,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jota"},{"type":"user_nicename","value":"jamiesho","user_id":32162,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=jamiesho"},{"type":"user_nicename","value":"cemke","user_id":31360,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=cemke"},{"type":"user_nicename","value":"shahrami","user_id":33590,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=shahrami"},{"type":"user_nicename","value":"awf","user_id":31157,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=awf"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[171411,238767],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":171411,"post_title":"Fully Articulated Hand Tracking","post_name":"fully-articulated-hand-tracking","post_type":"msr-project","post_date":"2014-10-02 20:03:22","post_modified":"2019-05-21 02:15:01","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/fully-articulated-hand-tracking\/","post_excerpt":"We present a new real-time articulated hand tracker which can enable new possibilities for human-computer interaction (HCI). Our system accurately reconstructs complex hand poses across a variety of subjects using only a single depth camera. It also allows for a high-degree of robustness, continually recovering from tracking failures. However, the most unique aspect of our tracker is its flexibility in terms of camera placement and operating range. &nbsp; Screenshots Please note, we are using a&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/171411"}]}},{"ID":238767,"post_title":"Subdivision Surfaces in Computer Vision","post_name":"subdivision-surfaces-in-computer-vision","post_type":"msr-project","post_date":"2016-10-05 05:54:25","post_modified":"2020-07-14 01:50:14","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/subdivision-surfaces-in-computer-vision\/","post_excerpt":"In vision and machine learning, almost everything we do may be considered to be a form of model fitting. Whether estimating the parameters of a convolutional neural network, computing structure and motion from image collections, tracking objects in video, computing low-dimensional representations of datasets, estimating parameters for an inference model such as Markov random fields, or extracting shape spaces such as active appearance models, it almost always boils down to minimizing an objective containing some&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/238767"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/168307"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/168307\/revisions"}],"predecessor-version":[{"id":524581,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/168307\/revisions\/524581"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=168307"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=168307"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=168307"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=168307"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=168307"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=168307"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=168307"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=168307"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=168307"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=168307"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=168307"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=168307"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=168307"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=168307"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=168307"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=168307"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}