[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82874-en":3,"doc-seo-82874-105":29,"detail-sidebar-cat-0-en-105":90},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":4,"is_deleted":4,"is_public":20,"is_downloadable":20,"audit_status":20,"page_count":21,"language":22,"language_code":23,"site_id":24,"html_lang":23,"table_of_contents":25,"faqs":26,"seo_title":13,"seo_description":14,"update_tm":27,"read_time":28},82874,8796095462418,"Noah","https://ap-avatar.wpscdn.com/avatar/80000253c1241d02b47?x-image-process=image/resize,m_fixed,w_180,h_180&k=1778826106357471780",8,"Research & Report","MemPose Category-level Object Pose Estimation with Memory","Robust and generalizable category-level object pose estimation is constrained by existing parametric approaches that encode category cues as fixed priors or static network weights, limiting scalability to highly diverse instances. MemPose reframes the task via a memory-centric design, adding an external memory buffer that stores and dynamically updates category-level geometric structural representations from previously observed objects. The method retrieves relevant patterns for current perception and fuses memory-derived information to predict 9-DoF object pose. Experiments on REAL275, CAMERA25, Housecat6D, and Wild6D validate superior performance.","arXiv :2607 .04930v 1 [ cs .CV] 6 Jul 2026  \nMemPose: Category-level Object Pose Estimation  \nwith Memory  \nXiao Lin 1 , Minghao Zhu 1 , Yun Peng 1 , Liuyi Wang 1 , Qiyi Wang 1 ,  \nChengju Liu 1 ,2 B , and Qijun Chen 1 ,2 B  \n1 Tongji University, Shanghai, China  \n2 State Key Laboratory of Autonomous Intelligent Unmanned Systems  \n{linx_xx, zmh_hh, pengyun, wly, wqy126179, liuchengju, [qjchen}@tongji.edu](qjchen}@tongji.edu), cn  \nAbstract. In the pursuit of robust and generalizable category-level object pose estimation, most existing methods adopt parametric formulations that learn effective representations from data, yet they primarily encode category-level patterns into fixed shape priors or static parameter weights, which limits their scalability to highly diverse instances.  \nIn this paper, we rethink category-level pose estimation from a memorycentric perspective and present MemPose, a memory-augmented framework that explicitly incorporates category-level geometric memory into the pose estimation pipeline. We introduce an external memory buffer that stores and dynamically updates structural representations from previously observed instances, enabling the model to leverage accumulated experience to support current perception. Extensive experiments on four challenging benchmarks (REAL275, CAMERA25, Housecat6D and Wild6D) demonstrate the superiority of our proposed method over previous state-of-the-art approaches.  \nKeywords: Category-level Pose Estimation · 3D Vision · Memory System  \n1 Introduction  \nAs a critical application in human-robot interaction [45], the Category-level Object Pose Estimation (COPE) [38] has attracted increasing attention. Unlike instance-level methods [22], COPE is model-free and aims to estimate the 9-DoF pose for arbitrary objects within predefined categories, without relying on instance-specific CAD models. However, this setting is inherently challenging due to the significant differences among objects within the category. To overcome these challenges, humans typically leverage memory from previous observations to perform analogical reasoning across instances. Such memory not only supports immediate perception but is also continuously updated as new objects are encountered.  \nDrawing on an intuitive understanding of human perception, a line of existing works [4, 18, 36] introduces explicit shape priors, often by extracting average  \n2 Xiao Lin, et al.  \nFig. 1: Overview of the category-level pose estimation pipeline: (a) Existing methods rely on static patterns, such as shape priors or fixed network parameters, to regress object pose and size. (b) In contrast, our approach introduces a dynamic, memoryaugmented pipeline that explicitly incorporates category-level geometric memory to enhance pose and size estimation.  \nshape for each category. Specifically, these approaches first reconstruct instance models by deforming a categorical shape prior and then match observations with the reconstructed models to regress pose. While priors provide explicit categorylevel cues, they are fixed once constructed and more like static prototypes. Essentially, such priors are hard to function as memory, as they would not update with new observations, thus capturing the diversity of instances. Moreover, acquiring the high-quality priors requires costly pre-processing pipelines.  \nMore recently, benefiting from the success of deep neural networks (DNNs), another line of existing methods adopt a parametric paradigm, which learns effective feature representations from input modalities via finely designed networks. For instance, HS-Pose [44] proposes a 3D graph convolution network to enhance pose-sensitive feature extraction, while AG-Pose [23] introduces a local feature aggregation module to establish robust keypoint-level correspondences. KeyPose [42] propose a graph-based detection method to strengthen the understanding of geometric structures. Furthermore, foundational models like DINOv2 [28] have been widely ado","cbCaiqlIlDgAo3Wz","https://ap.wps.com/l/cbCaiqlIlDgAo3Wz","pdf",1991058,1,18,"English","en",105,"# Introduction\n## Category-level Object Pose Estimation (COPE)\n## Limitations of Existing Methods\n## Memory-centric Proposal: MemPose","[{\"question\":\"What problem does MemPose address in category-level object pose estimation?\",\"answer\":\"MemPose targets the scalability limitation of methods that encode category knowledge with fixed shape priors or static network parameters, which cannot adapt to diverse instances during inference.\"},{\"question\":\"How does MemPose incorporate memory into the pose estimation pipeline?\",\"answer\":\"MemPose introduces an external memory buffer that stores and dynamically updates category-level geometric structural representations and retrieves relevant patterns to support current observations.\"},{\"question\":\"What pose output does MemPose predict and how is it computed?\",\"answer\":\"MemPose predicts the 9-DoF object pose by fusing memory-derived information with current features after updating the memory buffer using similarity-based token merge.\"}]",1784183598,45,{"code":4,"msg":30,"data":31},"ok",{"site_id":24,"language":23,"slug":32,"title":13,"keywords":33,"description":14,"schema_data":34,"social_meta":85,"head_meta":87,"extra_data":89,"updated_unix":27},"mempose-category-level-object-pose-estimation-with-memory","",{"@graph":35,"@context":84},[36,53,67],{"@type":37,"itemListElement":38},"BreadcrumbList",[39,43,47,50],{"item":40,"name":41,"@type":42,"position":20},"https://docshare.wps.com","Home","ListItem",{"item":44,"name":45,"@type":42,"position":46},"https://docshare.wps.com/document/","Document",2,{"item":48,"name":12,"@type":42,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":42,"position":52},"https://docshare.wps.com/document/mempose-category-level-object-pose-estimation-with-memory/82874/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":61,"encodingFormat":60,"isAccessibleForFree":62,"interactionStatistic":63},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-16",true,{"@type":64,"interactionType":65,"userInteractionCount":4},"InteractionCounter",{"@type":66},"ViewAction",{"@type":68,"mainEntity":69},"FAQPage",[70,76,80],{"name":71,"@type":72,"acceptedAnswer":73},"What problem does MemPose address in category-level object pose estimation?","Question",{"text":74,"@type":75},"MemPose targets the scalability limitation of methods that encode category knowledge with fixed shape priors or static network parameters, which cannot adapt to diverse instances during inference.","Answer",{"name":77,"@type":72,"acceptedAnswer":78},"How does MemPose incorporate memory into the pose estimation pipeline?",{"text":79,"@type":75},"MemPose introduces an external memory buffer that stores and dynamically updates category-level geometric structural representations and retrieves relevant patterns to support current observations.",{"name":81,"@type":72,"acceptedAnswer":82},"What pose output does MemPose predict and how is it computed?",{"text":83,"@type":75},"MemPose predicts the 9-DoF object pose by fusing memory-derived information with current features after updating the memory buffer using similarity-based token merge.","https://schema.org",{"og:url":51,"og:type":86,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":88,"canonical":51},"index,follow",{"doc_id":7,"site_id":24},{"code":4,"msg":5,"data":91},[92,96,100,104,109,114,119,122,127,130,134],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":93,"show_sort_weight":94,"slug":95},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":97,"show_sort_weight":98,"slug":99},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":101,"show_sort_weight":102,"slug":103},"Exam",70,"exam",{"id":105,"doc_module":4,"doc_module_name":45,"category_name":106,"show_sort_weight":107,"slug":108},5,"Comic",60,"comic",{"id":110,"doc_module":4,"doc_module_name":45,"category_name":111,"show_sort_weight":112,"slug":113},6,"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":45,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":45,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":45,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":45,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":45,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":45,"category_name":136,"show_sort_weight":105,"slug":137},19,"General","general"]