{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/occlusion-net-2d3d-occluded-keypoint","title":"Occlusion-Net: 2D/3D Occluded Keypoint Localization Using Graph Networks","arxiv_id":null,"date":"2019-06-01","proceeding":"CVPR 2019 6","authors":["N. Dinesh Reddy"," Minh Vo"," Srinivasa G. Narasimhan"],"abstract":"We present Occlusion-Net, a framework to predict 2D and 3D locations of occluded keypoints for objects, in a largely self-supervised manner. We use an off-the-shelf detector as input (like MaskRCNN) that is trained only on visible key point annotations. This is the only supervision used in this work. A graph encoder network then explicitly classifies invisible edges and a graph decoder network corrects the occluded keypoint locations from the initial detector. Central to this work is a trifocal tensor loss that provides indirect self-supervision for occluded keypoint locations that are visible in other views of the object. The 2D keypoints are then passed into a 3D graph network that estimates the 3D shape and camera pose using the self-supervised re-projection loss. At test time, our approach successfully localizes keypoints in a single view under a diverse set of severe occlusion settings. We demonstrate and evaluate our approach on synthetic CAD data as well as a large image set capturing vehicles at many busy city intersections. As an interesting aside, we compare the accuracy of human labels of invisible keypoints against those obtained from geometric trifocal-tensor loss.\r","url_abs":"http://openaccess.thecvf.com/content_CVPR_2019/html/Reddy_Occlusion-Net_2D3D_Occluded_Keypoint_Localization_Using_Graph_Networks_CVPR_2019_paper.html","url_pdf":"http://openaccess.thecvf.com/content_CVPR_2019/papers/Reddy_Occlusion-Net_2D3D_Occluded_Keypoint_Localization_Using_Graph_Networks_CVPR_2019_paper.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"occlusion-net-2d3d-occluded-keypoint","repo_url":"https://github.com/dineshreddy91/Occlusion_Net","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"3d-car-instance-understanding","task_name":"3D Car Instance Understanding"},{"task_slug":"3d-object-reconstruction-from-a-single-image","task_name":"3D Object Reconstruction From A Single Image"},{"task_slug":"3d-pose-estimation","task_name":"3D Pose Estimation"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"pose-estimation","task_name":"Pose Estimation"},{"task_slug":"vehicle-pose-estimation","task_name":"Vehicle Pose Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/3d-pose-estimation-on-carfusion","task":"3D Pose Estimation","dataset":"CarFusion","model":"Occlusion-NET","rank_in_archive_order":1,"of":1,"metrics":{"3DPCK":"93.2"},"uses_additional_data":false},{"leaderboard":"/sota/vehicle-pose-estimation-on-vehicle-pose","task":"Vehicle Pose Estimation","dataset":"CarFusion","model":"Occlusion-NET","rank_in_archive_order":1,"of":1,"metrics":{"PCK":"88.8"},"uses_additional_data":false}],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}