[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86095-en":3,"doc-seo-86095-105":30,"detail-sidebar-cat-0-en-105":91},{"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":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86095,2336464648746,"Skyler","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","EquiFusion: Kinematics-Agnostic Human Motion Prediction via Equivariant Latent Diffusion","EquiFusion proposes the first kinematics-agnostic stochastic 3D human motion prediction model to remove rigid skeleton hard-coding that blocks generalization. The method uses a latent diffusion framework with a permutation-equivariant architecture, treating kinematics connectivity as explicit input so computations are inherently agnostic to joint ordering and graph structure. This design enables truly cross-dataset training and zero-shot inference on unseen kinematics, including partial or occluded observations, while delivering state-of-the-art benchmark results and higher efficiency.","arXiv :2607 . 10984v1 [ cs .CV] 13 Jul 2026  \nEquiFusion: Kinematics-Agnostic Human Motion Prediction via Equivariant Latent Diffusion  \nCecilia Curreli 1 ,2 Abhishek Saroha 1 ,2  \nFlorian Hofherr 1 ,2 Riccardo Marin 1 ,2  \nDominik Muhle 1 ,2 Daniel Cremers 1 ,2  \n1 Technical University of Munich, Munich, Germany  \n2 Munich Center for Machine Learning, Munich, Germany  \nFig. 1: EquiFusion. We introduce the first model for stochastic human motion prediction that generalizes to unseen skeleton parameterization, i.e. kinematics. While previous methods require a trained instance for each dataset or better kinematics, with a single model we unlock training on multiple datasets and inference on motion parametrized with different kinematics. EquiFusion is the first SHMP model to handle zero-shot novel kinematics and occluded limbs without explicit training, while achieving state-of-the-art results on established benchmarks.  \nAbstract. Existing Stochastic 3D Human Motion Prediction models are fundamentally constrained by hard-coding the skeleton kinematics, severely limiting generalization, preventing cross-dataset training, and requiring complex data retargeting. We introduce EquiFusion, the first kinematics-agnostic model to solve this bottleneck, implementing a latent diffusion model with a permutation equivariant architecture. EquiFusion treats the kinematics’ connectivity as an explicit input parameter, ensuring its internal computations are inherently agnostic to joint ordering and graph structure. This novel design enables truly cross-dataset generalization to unseen kinematics and unlocks novel zero-shot directions, such as motion prediction from partial or occluded observations  \n2 C. Curreli et al.  \nand targeted limb generation. EquiFusion achieves state-of-the-art results on major benchmarks, being up to 75% more compact than previous kinematics-specific methods, while achieving faster training and inference. EquiFusion thus establishes a new, flexible standard for robust human motion prediction. Model and training code available on our project page.  \n1 Introduction  \nPredicting future human motion from past observations is a core component of human intelligence and a prerequisite for mature spatial AI. Due to the inherent ambiguity of human intent, the field has shifted from deterministic toward Stochastic Human Motion Prediction (SHMP), which models a distribution of diverse, physically plausible futures from an observation, with applications including human-robot collaboration, autonomous navigation, and augmented telepresence. While recent advances in generative models [13, 42 , 98] have further popularized SHMP probabilistic formulations [9, 18 , 23 , 108], a fundamental bottleneck remains: kinematics rigidity. Diverse datasets (e.g., Human3.6M [46], AMASS [76], and more recently Nymeria [75]) often employ different motioncapturing technologies, resulting in diverse skeleton parametrizations i. e. kinematics. Motion retargeting [8, 43 , 57], i.e., converting a motion to a different kinematics parametrization, is often not possible without introducing errors and drifts [66, 116 , 117] . The downstream impact on SHMP methods is tremendous. Previous works [9, 18 , 23 , 26 , 78 , 108 , 136] have hard-coded kinematics as a network structural prior, resulting in a plethora of skeleton-specific networks. Training a network for every different kinematics is impractical, inefficient, and does not generalize to new skeletal configurations.  \nWe break these dataset-specific boundaries by addressing settings involving heterogeneous kinematic structures. We propose two zero-shot kinematics inference tasks for SHMP: predicting motion for (i) full-body kinematics unseen at training time, and (ii) kinematics with occlusions or missing parts (i.e., partial observations) . Both these tasks address real-world use cases and important milestones toward mature spatial AI and foundation human motion models. To date, this has been unaddres","cbCaisHIO1DBlNdD","https://ap.wps.com/l/cbCaisHIO1DBlNdD","pdf",17652136,2,1,63,"English","en",105,"# Abstract\n# Introduction","[{\"question\":\"What problem does EquiFusion address in stochastic human motion prediction?\",\"answer\":\"EquiFusion targets the kinematics rigidity of prior SHMP models, which hard-code skeleton parameters and limit generalization and cross-dataset training. It also avoids the need for complex retargeting and dataset-specific network structures.\"},{\"question\":\"How does EquiFusion achieve kinematics-agnostic behavior?\",\"answer\":\"EquiFusion incorporates kinematics connectivity as an explicit input and uses a permutation-equivariant architecture. This makes the model’s internal computations inherently agnostic to joint ordering and graph structure.\"},{\"question\":\"What zero-shot capabilities does EquiFusion enable?\",\"answer\":\"EquiFusion supports zero-shot inference on full-body kinematics unseen during training and on kinematics with occlusions or missing parts from partial observations. It achieves state-of-the-art performance on established benchmarks.\"}]",1784208481,159,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"equifusion-kinematics-agnostic-human-motion-prediction-via-equivariant-latent-diffusion","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/equifusion-kinematics-agnostic-human-motion-prediction-via-equivariant-latent-diffusion/86095/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-26","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does EquiFusion address in stochastic human motion prediction?","Question",{"text":75,"@type":76},"EquiFusion targets the kinematics rigidity of prior SHMP models, which hard-code skeleton parameters and limit generalization and cross-dataset training. It also avoids the need for complex retargeting and dataset-specific network structures.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does EquiFusion achieve kinematics-agnostic behavior?",{"text":80,"@type":76},"EquiFusion incorporates kinematics connectivity as an explicit input and uses a permutation-equivariant architecture. This makes the model’s internal computations inherently agnostic to joint ordering and graph structure.",{"name":82,"@type":73,"acceptedAnswer":83},"What zero-shot capabilities does EquiFusion enable?",{"text":84,"@type":76},"EquiFusion supports zero-shot inference on full-body kinematics unseen during training and on kinematics with occlusions or missing parts from partial observations. It achieves state-of-the-art performance on established benchmarks.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]