Research map / 3D vision / 3D generation
Beyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis
Xianhao Chen, Jiarui Hu, Yuanbo Yang, Xiyu Zhang, Tengyue Wang, Hujun Bao, Guofeng Zhang, Zhaopeng Cui
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add scene graph message passing (edges) (ablation) | RelGraphOV · ScanNetV2 | mIoU: 50.7 -> 58.4 | Improved |
| add scene graph message passing (edges) (ablation) | RelGraphOV · ScanNet200 | mIoU: 9.5 -> 14.5 | Improved |
| add edge-feature (relation) guidance in attention (ablation) | RelGraphOV · ScanNetV2 | mIoU: 55.5 -> 58.4 | Improved |
| add edge-feature (relation) guidance in attention (ablation) | RelGraphOV · ScanNet200 | mIoU: 11.0 -> 14.5 | Improved |
| add global super-node context injection (ablation) | RelGraphOV · ScanNetV2 | mIoU: 56.0 -> 58.4 | Improved |
| add global super-node context injection (ablation) | RelGraphOV · ScanNet200 | mIoU: 11.1 -> 14.5 | Improved |
| use decoupled dual-stream instead of naive concatenation (ablation) | RelGraphOV · ScanNetV2 | mIoU: 57.9 -> 58.4 | Improved |
| use decoupled dual-stream instead of naive concatenation (ablation) | RelGraphOV · ScanNet200 | mIoU: 12.7 -> 14.5 | Improved |
| add hierarchical contrastive loss (ablation) | RelGraphOV · ScanNetV2 | mIoU: 52.2 -> 58.4 | Improved |
| add hierarchical contrastive loss (ablation) | RelGraphOV · ScanNet200 | mIoU: 7.8 -> 14.5 | Improved |
| add main-stream alignment loss L_reg (ablation) | RelGraphOV · ScanNetV2 | mIoU: 54.9 -> 58.4 | Improved |
| add auxiliary-stream alignment loss L_aux (ablation) | RelGraphOV · ScanNetV2 | mIoU: 55.9 -> 58.4 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- OpenScene — OpenScene: 3D Scene Understanding With Open Vocabularies (CVPR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- SceneScaffold: Active Scene-State Construction for Unified 3D Scene Understanding (NeurIPS 2026 · similarity 0.91)
- OpenLex3D: A Tiered Benchmark for Open-Vocabulary 3D Scene Representations (NeurIPS 2025 · similarity 0.91)
- Open3DSG: Open-Vocabulary 3D Scene Graphs from Point Clouds with Queryable Objects and Open-Set Relationships (CVPR 2024 · similarity 0.91)
- Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding (CVPR 2025 · similarity 0.91)
- A Unified Framework for 3D Scene Understanding (NeurIPS 2024 · similarity 0.91)
- Object-Centric Representation Learning for Enhanced 3D Semantic Scene Graph Prediction (NeurIPS 2025 · similarity 0.91)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
