Research map / 3D vision / 3D generation
Incremental 3D Semantic Scene Graph Prediction From RGB Sequences
Shun-Cheng Wu, Keisuke Tateno, Nassir Navab, Federico Tombari
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 |
|---|---|---|---|
| Use simple mean aggregation rather than CVR canonical transformation for multi-view features | ResNet18 object classification · ScanNet | mIoU (%): 32.3 -> 39.2 | Improved |
| Replace point-cloud-union edge features with relative pose descriptors (ablation) | Incremental scene graph method · 3RScan GT and Sparse; Dense | Rel. recall (%): 62.0 -> 66.1; 35.1 -> 34.1; 10.0 -> 9.9 | Mixed |
| Add a sigmoid gate to the geometric feature (ablation) | Incremental scene graph method · 3RScan GT | Rel. recall (%): 61.5 -> 66.1 | Improved |
| Add a sigmoid gate to the geometric feature (ablation) | Incremental scene graph method · 3RScan Dense | Rel. recall (%): 32.9 -> 34.1 | Improved |
| Add a sigmoid gate to the geometric feature (ablation) | Incremental scene graph method · 3RScan Sparse | Rel. recall (%): 8.6 -> 9.9 | Improved |
| Use the proposed method rather than the strongest baseline | 3RScan scene graph prediction · GT, without None relationship | Rel. recall (%): 49.0 -> 64.8 | Improved |
| Use the proposed method rather than the strongest baseline | 3RScan scene graph prediction · Dense, without None relationship | Rel. recall (%): 24.4 -> 25.5 | Improved |
| Use the proposed method rather than the strongest baseline | 3RScan scene graph prediction · Sparse, without None relationship | Rel. recall (%): 3.7 -> 6.9 | Improved |
Closest papers by meaning
- SceneGraphFusion: Incremental 3D Scene Graph Prediction From RGB-D Sequences (CVPR 2021 · similarity 0.92)
- Learning 3D Semantic Scene Graphs with Instance Embeddings (IJCV 2022 · similarity 0.91)
- Statistical Confidence Rescoring for Robust 3D Scene Graph Generation from Multi-View Images (ICCV 2025 · similarity 0.91)
- Object-Centric Representation Learning for Enhanced 3D Semantic Scene Graph Prediction (NeurIPS 2025 · similarity 0.89)
- Open3DSG: Open-Vocabulary 3D Scene Graphs from Point Clouds with Queryable Objects and Open-Set Relationships (CVPR 2024 · similarity 0.88)
- DSGG: Dense Relation Transformer for an End-to-end Scene Graph Generation (CVPR 2024 · similarity 0.88)
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.
