[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-85212-en":3,"doc-seo-85212-105":30,"detail-sidebar-cat-0-en-105":83},{"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},85212,13056703019662,"Evangeline","https://ap-avatar.wpscdn.com/avatar/be000253a8e92610077?_k=1778726343310543188",8,"Research & Report","GNOCHI Generative Neural Model for Close Human-Human Interactions","Creating realistic 3D human-human interactions in virtual environments is difficult because articulated bodies have many degrees of freedom and interactions must remain physically accurate without interpenetration or mesh collisions. Existing tracking- or reconstruction-based approaches lack generative sampling, while current text/image generative methods often fail to model close contacts. This work presents GNOCHI, a conditional variational autoencoder that generates a reaction human pose conditioned on another human’s pose, enabling controlled and diverse interaction synthesis.","ACM SIGGRAPH / Eurographics Symposium on Computer Animation 2026 D. Levin and M. Chu  \n(Guest Editors)  \nCOMPUTER GRAPHICS forum Volume 45 (2026), Number 8  \nGNOCHI: Generative Neural mOdel for Close Human-Human  \nInteractions  \nGonzalo Gómez-Nogales†1 Marc Comino-Trinidad†1 Andrés Casado-Elvira 1 Dan Casas‡1  \n1Universidad Rey Juan Carlos, Madrid, Spain.  \narXiv :2607 . 10408v1 [ cs .CV] 11 Jul 2026  \nFigure 1: Given an input 3D pose (i.e., the conditioning pose, in purple ), our generative model infers a 3D pose of a human in close interaction (i.e., the reaction pose, in green). This enables the conditional generation of 3D humans in close interaction, which can be used fine-grain control signal for image generative methods (right) .  \nAbstract  \nCreating realistic 3D human-human interactions in virtual environments is challenging due to the high degrees of freedom inhuman body and the need for physically accurate poses that do not collide with each other. Traditional methods for humanhuman interaction are based on motion tracking or 3D body reconstruction, but lack generative capabilities. Recent generative methods enable the synthesis of individual or interacting motions via text or image input, but generally fall short in modeling close interactions. This paper introduces a novel generative model for close 3D human-human interactions using a conditional  \nvariational autoencoder (cVAE), which generates poses for one human conditioned on the pose of another, allowing for con trolled and diverse interaction synthesis. To train our model, we address two underlying long-standing challenges in the field of human-human interaction: data scarcity, for which we propose an automated supervised data augmentation strategy that generates synthetic yet realistic interaction poses; and collision awareness in generative approaches, for which we propose a self-supervised loss based on a collision resolution technique using volumetric proxies to ensure physically correct interac   \ntions. We extensively evaluate the capabilities of our model, and demonstrate a wide variety of plausible and physically correct interactions, not possible to generate with current state-of-the-art methods.  \nCCS Concepts  \n• Computing methodologies → Physical simulation; Collision detection; Mesh models;  \n1. Introduction  \nCreating life-like virtual 3D scenes is central to Computer Vision, Graphics, and VR, impacting dataset generation, human track-  \n† Equal contribution  \n‡ Work done prior to joining Amazon  \ning, 3D scene understanding, and character animation. However, modeling and reconstructing how humans interact in everyday 3D scenes is a highly complex task due to the large number of degrees of freedom involved. Additionally, as humans, we are very sensitive to non-physically-correct virtual scene arrangements (e.g., interpenetrations, mesh collisions, or impossible configurations), hence,  \n© 2026 The Authors. Computer Graphics Forum published by Eurographics-The European Association for Computer Graphics and John Wiley & Sons Ltd.  \nThis is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.  \n2 of 13 Gómez-Nogales et al. / GNOCHI: Generative Neural mOdel for Close Human-Human Interactions  \nerrors in modeling human interactions automatically produce significant visual discomfort.  \nExisting research on 3D interaction generally falls into three categories: human-scene methods [JZL∗24, HGT∗21, GCM∗24, ZZW∗23,ZWZ∗22, HCV∗21,MCZ∗25], for large-scale navigation; human-object methods [JLC∗23, XBPM22, CKA∗22, HVT∗19, YGKT24, TCBT22], for manipulation and grasping; and humanhuman models [MYP∗24, YPMK23, FZO∗20, FZS∗21, MMR∗24] . We focus on the latter, specifically addressing the synthesis of natural, close-contact interactions between two highly articulated bodies.  \nUnfortunately, most existing human-human intera","cbCaipB3qcY4mU40","https://ap.wps.com/l/cbCaipB3qcY4mU40","pdf",9903487,3,1,13,"English","en",105,"# Abstract\n# Introduction\n## Motivation and challenges\n## Related work and limitations","[{\"question\":\"How does the method handle physical collision awareness during generation?\",\"answer\":\"The paper introduces a self-supervised collision-aware loss using volumetric proxies and a collision resolution technique to ensure physically correct interactions during generative training and sampling.\"}]",1784201781,33,{"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":78,"head_meta":80,"extra_data":82,"updated_unix":28},"gnochi-generative-neural-model-for-close-human-human-interactions","",{"@graph":36,"@context":77},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/gnochi-generative-neural-model-for-close-human-human-interactions/85212/",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","2026-07-20","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71],{"name":72,"@type":73,"acceptedAnswer":74},"How does the method handle physical collision awareness during generation?","Question",{"text":75,"@type":76},"The paper introduces a self-supervised collision-aware loss using volumetric proxies and a collision resolution technique to ensure physically correct interactions during generative training and sampling.","Answer","https://schema.org",{"og:url":51,"og:type":79,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":81,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":84},[85,89,93,97,102,107,112,115,120,123,127],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":86,"show_sort_weight":87,"slug":88},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":90,"show_sort_weight":91,"slug":92},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Exam",70,"exam",{"id":98,"doc_module":4,"doc_module_name":46,"category_name":99,"show_sort_weight":100,"slug":101},5,"Comic",60,"comic",{"id":103,"doc_module":4,"doc_module_name":46,"category_name":104,"show_sort_weight":105,"slug":106},6,"Technology",50,"technology",{"id":108,"doc_module":4,"doc_module_name":46,"category_name":109,"show_sort_weight":110,"slug":111},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":113,"slug":114},30,"research-report",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":46,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":98,"slug":130},19,"General","general"]