[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-82936-en":3,"doc-seo-82936-105":29,"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":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},82936,8796095461610,"Oliver","https://ap-avatar.wpscdn.com/davatar_276721f389ce27ea32af1340a28f341c",8,"Research & Report","Beyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis","Open-vocabulary 3D scene understanding targets dense semantic segmentation that goes beyond fixed category sets by transferring semantics from vision-language models. Existing work lifts 2D language-aligned features into 3D but often uses context-independent representations, leaving object relationships insufficiently exploited. RelGraphOV introduces relationship-aware contextual refinement through 3D scene graphs, inferring object relations from multi-view observations with vision-language reasoning and pruning implausible edges. An Adaptive Gated Dual-Stream Contextual GAT with hierarchical contrastive learning improves instance consistency and category discrimination, validated on multiple benchmarks.","arXiv :2607 .05348v 1 [ cs .CV] 6 Jul 2026  \nBeyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis  \nXianhao Chen 1 ∗, Jiarui Hu 1 ∗, Yuanbo Yang2, Xiyu Zhang 1, Tengyue Wang2, Hujun Bao 1, Guofeng Zhang 1, and Zhaopeng Cui 1†  \n1 State Key Lab of CAD&CG, Zhejiang University, China  \n2 Zhejiang University, China  \nAbstract. Open-vocabulary 3D scene understanding aims to segment 3D scenes beyond predefined categories by transferring semantic knowledge from vision-language models. Existing methods have advanced this task by lifting language-aligned 2D features into 3D, yet they often rely on context-independent semantic representations, leaving object relationships underexplored for contextual refinement. We propose RelGraphOV, a relationship-aware framework that uses 3D scene graphs to enhance open-vocabulary 3D understanding. Our method constructs relational scene graphs from multi-view observations by leveraging visionlanguage reasoning to infer object relationships and prune geometrically implausible connections, without manual relationship annotations. To aggregate relational context while avoiding feature interference, we introduce an Adaptive Gated Dual-Stream Contextual GAT that separates dense geometric features and semantic CLIP embeddings, performs edgeguided message passing, and adaptively fuses complementary semantics.  \nA hierarchical contrastive objective further promotes instance-level consistency and category-level discrimination. Experiments on ScanNetV2, ScanNet200, ScanNet++, and Replica demonstrate strong performance and generalization ability. Project Page: [cxavireh.github.io/relgraphov](cxavireh.github.io/relgraphov)projectpage  \nKeywords: 3D Semantic Segmentation · Scene Graph · Open-Vocabulary  \n1 Introduction  \nOpen-vocabulary 3D scene understanding primarily aims to perform dense semantic segmentation of 3D environments, extending object recognition beyond a fixed set of predefined categories to enable scalable perception for applications such as robotic navigation, spatial reasoning, and augmented reality. Inspired by vision-language models such as CLIP [35], recent works transfer semantic knowledge learned from image-text data into 3D representations. Existing open-vocabulary 3D semantic segmentation approaches generally fall into two  \n∗ Equal contribution. † Corresponding author.  \n2 Chen et al.  \nRelationship-aware Scene Graph Illustration  \nFig. 1: RelGraphOV. We propose a relationship-aware open-vocabulary 3D scene understanding framework. Left: Illustration of encoding a 3D scene into a relationshipaware scene graph to capture contextual relationships. Right: Qualitative openvocabulary segmentation results produced by our method.  \nparadigms. The first projects dense vision-language features from 2D models into 3D space [25, 33], which provides strong spatial robustness under occlusion or incomplete observations. The second associates instance-level CLIP embeddings with 3D object proposals [40, 44, 48], enabling stronger semantic generalization to long-tail categories. Despite their complementary strengths, both paradigms primarily rely on context-independent semantic representations, leaving structured object relationships underexplored for semantic refinement. For instance, distinguishing a “curtain” from a “shower curtain” is notoriously difficult when relying exclusively on isolated object appearances. However, if the model is aware of the surrounding context, such as the object’s spatial relationship to a bathtub or a toilet, this semantic ambiguity can be naturally resolved.  \nA natural representation for modeling such contextual information is the 3D scene graph, which represents objects as nodes and their relationships as edges. Existing relational 3D methods have studied scene graph prediction [19, 20, 42, 45,47], relationship querying [12,21], referred object grounding [3], and relationaware instance segmentation [29] . However, t","cbCairhryF9nJuPF","https://ap.wps.com/l/cbCairhryF9nJuPF","pdf",8295900,1,20,"English","en",105,"# Introduction\n## Open-vocabulary 3D scene understanding\n## Limitations of context-independent representations\n## Motivation for 3D scene graphs\n# Proposed Method\n## RelGraphOV framework overview\n## Scene graph construction without manual annotations\n## Adaptive gated dual-stream contextual GAT\n# Training Objective and Evaluation\n## Hierarchical contrastive objective\n## Experimental results and generalization","[{\"question\":\"What problem does RelGraphOV address in open-vocabulary 3D scene understanding?\",\"answer\":\"RelGraphOV focuses on improving open-vocabulary dense 3D segmentation by exploiting object relationships, addressing the limitation that prior methods often rely on context-independent representations.\"},{\"question\":\"How does the method obtain relationships without manual relationship annotations?\",\"answer\":\"It constructs a 3D scene graph from multi-view observations using vision-language reasoning to infer object relationships, then prunes geometrically implausible connections.\"},{\"question\":\"What mechanism helps avoid feature interference when combining geometric and semantic information?\",\"answer\":\"It introduces an Adaptive Gated Dual-Stream Contextual GAT that separates dense geometric features and semantic CLIP embeddings, performs edge-guided message passing, and adaptively fuses complementary semantics.\"}]",1784184138,50,{"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":86,"head_meta":88,"extra_data":90,"updated_unix":27},"beyond-isolated-objects-relationship-aware-open-vocabulary-scene-understanding-via-3d-scene-graph-analysis","",{"@graph":35,"@context":85},[36,53,68],{"@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/beyond-isolated-objects-relationship-aware-open-vocabulary-scene-understanding-via-3d-scene-graph-analysis/82936/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":23,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":40,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-17","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 RelGraphOV address in open-vocabulary 3D scene understanding?","Question",{"text":75,"@type":76},"RelGraphOV focuses on improving open-vocabulary dense 3D segmentation by exploiting object relationships, addressing the limitation that prior methods often rely on context-independent representations.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the method obtain relationships without manual relationship annotations?",{"text":80,"@type":76},"It constructs a 3D scene graph from multi-view observations using vision-language reasoning to infer object relationships, then prunes geometrically implausible connections.",{"name":82,"@type":73,"acceptedAnswer":83},"What mechanism helps avoid feature interference when combining geometric and semantic information?",{"text":84,"@type":76},"It introduces an Adaptive Gated Dual-Stream Contextual GAT that separates dense geometric features and semantic CLIP embeddings, performs edge-guided message passing, and adaptively fuses complementary semantics.","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":24},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,126,129,133],{"id":20,"doc_module":4,"doc_module_name":45,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":46,"doc_module":4,"doc_module_name":45,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":45,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":45,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":45,"category_name":112,"show_sort_weight":28,"slug":113},6,"Technology","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":21,"slug":125},9,"Religion & Spirituality","religion-spirituality",{"id":21,"doc_module":4,"doc_module_name":45,"category_name":127,"show_sort_weight":21,"slug":128},"World Cup","world-cup",{"id":130,"doc_module":4,"doc_module_name":45,"category_name":131,"show_sort_weight":130,"slug":132},10,"Lifestyle","lifestyle",{"id":134,"doc_module":4,"doc_module_name":45,"category_name":135,"show_sort_weight":106,"slug":136},19,"General","general"]