Research map / 3D vision / 3D generation
ViewSRD: 3D Visual Grounding via Structured Multi-View Decomposition
Ronggang Huang, Haoxin Yang, Yan Cai, Xuemiao Xu, Huaidong Zhang, Shengfeng He
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 CCVTs (ablation) | ViewSRD · Nr3D | overall accuracy: 62.2 -> 69.9 | Improved |
| add scene module (ablation) | ViewSRD · Nr3D | overall accuracy: 64.6 -> 69.9 | Improved |
| add textual module (ablation) | ViewSRD · Nr3D | overall accuracy: 68.0 -> 69.9 | Improved |
| add SRD module (ablation) | ViewSRD · Nr3D | overall accuracy: 68.6 -> 69.9 | Improved |
| add view-alignment weighting (ablation) | ViewSRD · Nr3D | overall accuracy: 69.0 -> 69.9 | Improved |
| add SRD decoupling to MVT | MVT · ScanRefer | Acc@0.5: 33.3 -> 34.3 | Improved |
| add SRD decoupling to BUTD-DETR | BUTD-DETR · ScanRefer | Acc@0.25: 50.4 -> 57.9 | Improved |
| add SRD decoupling to EDA | EDA · ScanRefer | Acc@0.5: 32.3 -> 35.3 | Improved |
| test with 4 views instead of 1 (trained with 4) (ablation) | ViewSRD · Nr3D | overall accuracy: 66.0 -> 69.9 | Improved |
| train and test with 8 views instead of 4 (ablation) | ViewSRD · Nr3D | overall accuracy: 69.9 -> 68.4 | Hurt |
| train and test with 4 views instead of 1 (ablation) | ViewSRD · Nr3D | overall accuracy: 64.4 -> 69.9 | Improved |
| swap DeepSeek-R1 for Qwen-Turbo as LLM decoupler (ablation) | SRD module · Nr3D | accuracy: 69.9% -> 70.7% | 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.
- MiKASA — MiKASA: Multi-Key-Anchor & Scene-Aware Transformer for 3D Visual Grounding (CVPR 2024) · 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
- SeeGround: See and Ground for Zero-Shot Open-Vocabulary 3D Visual Grounding (CVPR 2025 · similarity 0.90)
- From Objects to Anywhere: A Holistic Benchmark for Multi-level Visual Grounding in 3D Scenes (NeurIPS 2025 · similarity 0.90)
- CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding (ICLR 2024 · similarity 0.90)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.89)
- S$^2$-MLLM: Boosting Spatial Reasoning Capability of MLLMs for 3D Visual Grounding with Structural Guidance (CVPR 2026 · similarity 0.89)
- Grounded 3D-Aware Spatial Vision-Language Modeling (CVPR 2026 · similarity 0.89)
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.
