[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"detail-sidebar-cat-0-en-105":3,"doc-seo-517159-105":59,"doc-detail-517159-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","hybridpose-6d-object-pose-estimation-under-hybrid-representations","HybridPose: 6D Object Pose Estimation under Hybrid Representations","","HybridPose presents a novel 6D object pose estimation method that uses a hybrid intermediate representation to capture multiple geometric cues from an RGB image: 2D keypoints, edge vectors, and symmetry correspondences. By combining these representations, pose regression can exploit richer and more diverse information than a single representation, especially when predictions become inaccurate due to occlusion. A shared neural network predicts all intermediate signals, while a robust regression module filters outliers. On Occlusion Linemod, HybridPose reaches 79.2% mean ADD(-S) at 30 fps, improving the state of the art by 67.4% while maintaining comparable runtime.",{"@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/hybridpose-6d-object-pose-estimation-under-hybrid-representations/517159/",{"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/hybridpose-6d-object-pose-estimation-under-hybrid-representations/517159.png","ImageObject",300,407,{"name":92,"@type":93},"Mafia Boss","Person",{"url":74,"name":95,"@type":96},"DocShare","Organization","application/pdf","2026-11-16","2026-10-07",true,{"@type":102,"interactionType":103,"userInteractionCount":105},"InteractionCounter",{"@type":104},"ViewAction",12,{"@type":107,"mainEntity":108},"FAQPage",[109,115,119],{"name":110,"@type":111,"acceptedAnswer":112},"What is HybridPose’s core idea for 6D object pose estimation?","Question",{"text":113,"@type":114},"HybridPose uses a hybrid intermediate representation that predicts keypoints, edge vectors, and symmetry correspondences from the input image, enabling pose regression to rely on multiple geometric cues.","Answer",{"name":116,"@type":111,"acceptedAnswer":117},"How does HybridPose address failures caused by occlusion or inaccurate intermediate predictions?",{"text":118,"@type":114},"It combines constraints from several intermediate representations and uses a robust regression module with trainable robust norms to prune outliers in predicted intermediate signals.",{"name":120,"@type":111,"acceptedAnswer":121},"What performance does HybridPose achieve on the Occlusion Linemod benchmark?",{"text":122,"@type":114},"HybridPose achieves 79.2% mean ADD(-S) accuracy while running at 30 fps, corresponding to a 67.4% improvement over the current state of the art mentioned in the text (DPOD).","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},517159,1791513085,{"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":52,"language":140,"language_code":63,"site_id":62,"html_lang":63,"table_of_contents":141,"faqs":142,"seo_title":143,"seo_description":67,"update_tm":144,"read_time":145,"preview_page_count":19,"preview_urls":146},962090760730,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","HybridPose: 6D Object Pose Estimation under Hybrid Representations  \nChen Song*, Jiaru Song􀀃 , Qixing Huang  \nThe University of Texas at Austin  \n[song@cs.utexas.edu](song@cs.utexas.edu), [jiarus@cs.utexas.edu](jiarus@cs.utexas.edu), [huangqx@cs.utexas.edu](huangqx@cs.utexas.edu)  \nAbstract  \nWe introduce HybridPose, a novel 6D object pose estimation approach. HybridPose utilizes a hybrid intermediate representation to express different geometric information in the input image, including keypoints, edge vectors, and symmetry correspondences. Compared to a unitary representation, our hybrid representation allows pose regression to exploit more and diverse features when one typeof predicted representation is inaccurate (e.g., because of occlusion). Different intermediate representations used by HybridPose can all be predicted by the same simple neural network, and outliers in predicted intermediate representations are 􀀂ltered by a robust regression module. Compared to state-of-the-art pose estimation approaches, HybridPose is comparable in running time and is signi􀀂cantly more accurate. For example, on Occlusion Linemod [3] dataset, our method achieves a prediction speed of 30 fps with a mean ADD(-S) accuracy of 79.2%, representing a 67.4% improvement from the current state-of-the-art approach. The implementation of HybridPose is available at [https://github.com/chensong1995/HybridPose](https://github.com/chensong1995/HybridPose).  \n1. Introduction  \nEstimating the 6D pose of an object from an RGB image is a fundamental problem in 3D vision and has diverse applications in object recognition and robot-object interaction. Advances in deep learning have led to signi􀀂cant breakthroughs in this problem. While early works typically formulate pose estimation as end-to-end pose classi􀀂 -cation [39] or pose regression [16, 42], recent pose estimation methods usually leverage keypoints as an intermediate representation [38, 34], and align predicted 2D keypoints with ground-truth 3D keypoints. In addition to ground-truth pose labels, these methods incorporate keypoints as an intermediate supervision, facilitating smooth model training. Keypoint-based methods are built upon two assumptions:  \n(1) a machine learning model can accurately predict 2Dkeypoint locations; and (2) these predictions provide suf-  \n* Authors contributed equally  \n(a) Input image (b) Keypoints (c) Edge vectors  \n(d) Symmetry correspondences (e) Groud-truth pose (f) Predicted pose  \nFigure 1 . HybridPose predicts keypoints, edge vectors, and symmetry correspondences. In (a), we show the input RGB image, in which the object of interest (driller) is partially occluded. In (b), red markers denote predicted 2D keypoints. In (c), edge vectors are de􀀂ned by a fully-connected graph among all keypoints. In (d), symmetry correspondences connect each 2D pixel on the object to its symmetric counterpart. For illustrative purposes, we only draw symmetry correspondences of 50 random samples from 5755 detected object pixels in this example. The predicted pose (f) is obtained by jointly aligning all predictions with the 3D template, which involves solving a non-linear optimization problem.  \n􀀂cient constraints to regress the underlying 6D pose. Both assumptions easily break in many real-world settings. Due to object occlusions and representational limitations of the prediction network, it is often impossible to accurately predict 2D keypoint coordinates from an RGB image alone.  \nIn this paper, we introduce HybridPose, a novel 6D pose estimation approach that leverages multiple intermediate representations to express the geometric information in the input image. In addition to keypoints, HybridPose integrates a prediction network that outputs edge vectors between adjacent keypoints. As most objects possess a (partial) re􀀃ection symmetry, HybridPose also utilizes predicted dense pixel-wise correspondences that re􀀃ect the underlying symmetric relations between pixels. Compared to a unit","cbCaiuCmH3ekasAQ","https://ap.wps.com/l/cbCaiuCmH3ekasAQ","pdf",2103776,"English","# Abstract\n# 1. Introduction\n## Problem: 6D pose estimation from RGB images\n## Keypoint-based intermediate representations and their assumptions\n## HybridPose: hybrid intermediate representations (keypoints, edge vectors, symmetry correspondences)\n## Robust pose regression and outlier pruning\n## Evaluation on Linemod and Occlusion Linemod","[{\"question\":\"What is HybridPose’s core idea for 6D object pose estimation?\",\"answer\":\"HybridPose uses a hybrid intermediate representation that predicts keypoints, edge vectors, and symmetry correspondences from the input image, enabling pose regression to rely on multiple geometric cues.\"},{\"question\":\"How does HybridPose address failures caused by occlusion or inaccurate intermediate predictions?\",\"answer\":\"It combines constraints from several intermediate representations and uses a robust regression module with trainable robust norms to prune outliers in predicted intermediate signals.\"},{\"question\":\"What performance does HybridPose achieve on the Occlusion Linemod benchmark?\",\"answer\":\"HybridPose achieves 79.2% mean ADD(-S) accuracy while running at 30 fps, corresponding to a 67.4% improvement over the current state of the art mentioned in the text (DPOD).\"}]","HybridPose: 6D Object Pose Estimation under Hybrid Representations | PDF",1791336248,25,[147,151,155,159],{"page_no":8,"url":148,"width":149,"height":150},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/3104bbf5d835e0b9365b9603edb36043.webp",983,1272,{"page_no":14,"url":152,"width":153,"height":154},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/6c9b2bc42183ba5dac0385df86f8f25e.webp",982,1270,{"page_no":81,"url":156,"width":157,"height":158},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/e90f21f62f384a0542fb16485a986be4.webp",981,1269,{"page_no":19,"url":160,"width":161,"height":162},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-17/7ebeacfaded29ff325a0998f7d10f9fb.webp",980,1268]