{"id":666252,"date":"2020-06-12T14:12:33","date_gmt":"2020-06-12T21:12:33","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/?post_type=msr-research-item&p=666252"},"modified":"2020-06-14T09:15:58","modified_gmt":"2020-06-14T16:15:58","slug":"higherhrnet-scale-aware-representation-learning-for-bottom-up-human-pose-estimation","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/higherhrnet-scale-aware-representation-learning-for-bottom-up-human-pose-estimation\/","title":{"rendered":"HigherHRNet: Scale-Aware Representation Learning for Bottom-Up Human Pose Estimation"},"content":{"rendered":"

Bottom-up human pose estimation methods have difficulties in predicting the correct pose for small persons due to challenges in scale variation. In this paper, we present HigherHRNet<\/strong>: a novel bottom-up human pose estimation method for learning scale-aware representations using high-resolution feature pyramids. Equipped with multi-resolution supervision for training and multiresolution aggregation for inference, the proposed approach is able to solve the scale variation challenge in bottom-up multi-person pose estimation and localize keypoints more precisely, especially for small person. The feature pyramid in HigherHRNet consists of feature map outputs from HRNet and upsampled higher-resolution outputs through a transposed convolution. HigherHRNet outperforms the previous best bottom-up method by 2.5% AP for medium person on COCO test-dev, showing its effectiveness in handling scale variation. Furthermore, HigherHRNet achieves new state-of-the-art result on COCO test-dev (70.5% AP) without using refinement or other post-processing techniques, surpassing all existing bottom-up methods. HigherHRNet even surpasses all topdown methods on CrowdPose test (67.6% AP), suggesting its robustness in crowded scene. The code and models are available at https:\/\/github.com\/HRNet\/Higher-HRNet-Human-Pose-Estimation (opens in new tab)<\/span><\/a>.<\/p>\n","protected":false},"excerpt":{"rendered":"

Bottom-up human pose estimation methods have difficulties in predicting the correct pose for small persons due to challenges in scale variation. In this paper, we present HigherHRNet: a novel bottom-up human pose estimation method for learning scale-aware representations using high-resolution feature pyramids. Equipped with multi-resolution supervision for training and multiresolution aggregation for inference, the proposed […]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13562],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-666252","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-computer-vision","msr-locale-en_us"],"msr_publishername":"","msr_edition":"","msr_affiliation":"","msr_published_date":"2020-3-12","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":"","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"url","viewUrl":"false","id":"false","title":"https:\/\/arxiv.org\/abs\/1908.10357","label_id":"243109","label":0},{"type":"file","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/06\/1908.10357.pdf","id":"666669","title":"1908-10357","label_id":"243132","label":0}],"msr_related_uploader":"","msr_attachments":[{"id":666669,"url":"https:\/\/www.microsoft.com\/en-us\/research\/uploads\/prod\/2020\/06\/1908.10357.pdf"}],"msr-author-ordering":[{"type":"text","value":"Bowen Cheng","user_id":0,"rest_url":false},{"type":"text","value":"Bin Xiao","user_id":0,"rest_url":false},{"type":"user_nicename","value":"Jingdong Wang","user_id":32299,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=Jingdong Wang"},{"type":"text","value":"Honghui Shi","user_id":0,"rest_url":false},{"type":"text","value":"Thomas S. Huang","user_id":0,"rest_url":false},{"type":"text","value":"Lei Zhang","user_id":0,"rest_url":false}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/666252"}],"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\/666252\/revisions"}],"predecessor-version":[{"id":666255,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/666252\/revisions\/666255"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=666252"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=666252"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=666252"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=666252"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=666252"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=666252"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=666252"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=666252"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=666252"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=666252"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=666252"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=666252"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=666252"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=666252"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=666252"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}