Research map / 3D vision / 3D generation
3D Concept Grounding on Neural Fields
Yining Hong, Yilun Du, Chunru Lin, Joshua B. Tenenbaum, Chuang Gan
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 |
|---|---|---|---|
| replace MAC with 3D-CG | MAC · PartNet-Reasoning Chair count_part | accuracy: 78.1 -> 92.2 | Improved |
| replace CVX+L with 3D-CG | CVX+L · PartNet-Reasoning Table count_part | accuracy: 30.7 -> 90.9 | Improved |
| replace CVX+L with 3D-CG | CVX+L · PartNet-Reasoning Chair query_part | accuracy: 72.3 -> 68.7 | Hurt |
| replace MAC with 3D-CG | MAC · PartNet-Reasoning Cart count_part | accuracy: 82.1 -> 74.8 | Hurt |
| replace NDF+LSTM with 3D-CG | NDF+LSTM · PartNet-Reasoning Bag count_part | accuracy: 72.3 -> 68.3 | Hurt |
| add question-answering loss to CVX (CVX+L) | CVX · PartNet-Reasoning semantic segmentation | mean IoU: 61.7 -> 65.2 | Improved |
| add question-answering loss to BAE (BAE+L) | BAE · PartNet-Reasoning semantic segmentation | mean IoU: 58.6 -> 61.3 | Improved |
| add question-answering loss to CVX (CVX+L) | CVX · PartNet-Reasoning instance segmentation | mean IoU: 34.6 -> 38.9 | Improved |
| replace CVX+L with 3D-CG | CVX+L · PartNet-Reasoning semantic segmentation | mean IoU: 65.2 -> 69.4 | Improved |
| replace BAE+L with 3D-CG | BAE+L · PartNet-Reasoning instance segmentation | mean IoU: 42.0 -> 56.9 | Improved |
| replace CVX+L with 3D-CG | CVX+L · PartNet-Reasoning Bag semantic segmentation | IoU: 70.2 -> 67.3 | Hurt |
Closest papers by meaning
- Naturally Supervised 3D Visual Grounding with Language-Regularized Concept Learners (CVPR 2024 · similarity 0.90)
- Grounded 3D-Aware Spatial Vision-Language Modeling (CVPR 2026 · similarity 0.88)
- Reasoning in Space via Grounding in the World (ICLR 2026 · similarity 0.88)
- Language-Image Models with 3D Understanding (ICLR 2025 · similarity 0.88)
- G^3-LQ: Marrying Hyperbolic Alignment with Explicit Semantic-Geometric Modeling for 3D Visual Grounding (CVPR 2024 · similarity 0.88)
- CoT3DRef: Chain-of-Thoughts Data-Efficient 3D Visual Grounding (ICLR 2024 · similarity 0.87)
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.
