Research map / 3D vision / 3D generation
Ov3R: Open-Vocabulary Semantic 3D Reconstruction from RGB Videos
Ziren Gong, Xiaohan Li, Fabio Tosi, Jiawei Han, Stefano Mattoccia, Jianfei Cai, Matteo Poggi
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 |
|---|---|---|---|
| insert object-level CLIP features into I2P (ablation) | CLIP3R · Replica | Accuracy (cm): 3.31 -> 3.05 | Improved |
| add CLIP-semantic supervision head in L2W (ablation) | CLIP3R · Replica | Accuracy (cm): 3.20 -> 3.05 | Improved |
| use object-level CLIP features instead of vanilla CLIP (ablation) | CLIP3R · Replica | Accuracy (cm): 3.18 -> 3.05 | Improved |
| insert CLIP features in I2P (ablation) | CLIP3R · Replica | ATE RMSE: 6.46 -> 6.00 | Improved |
| add DINO features to the fused descriptor (ablation) | 2D-3D OVS · Replica 3D segmentation | mIoU: 28.05 -> 30.65 | Improved |
| add 3D-CLIP encoder features to the fused descriptor (ablation) | 2D-3D OVS · Replica 3D segmentation | mIoU: 28.46 -> 30.65 | Improved |
| run 2D-3D OVS on CLIP3R reconstruction instead of OVO CLIP merging | Ov3R · Replica 3D segmentation (CLIP3R geometry) | mIoU: 25.6 -> 30.4 | Improved |
| Ov3R 2D-3D OVS vs OpenScene on ScanNet20 with GT geometry | Ov3R · ScanNet20 (GT geometry) | mIoU: 44.6 -> 43.6 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- SLAM3R — SLAM3R: Real-Time Dense Scene Reconstruction from Monocular RGB Videos (CVPR 2025) · better on 5 of 5 shared results
- DUSt3R — DUSt3R: Geometric 3D Vision Made Easy (CVPR 2024) · better on 1 of 1 shared result
- OpenScene — OpenScene: 3D Scene Understanding With Open Vocabularies (CVPR 2023) · better on 0 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
- Open-Vocabulary 3D Semantic Segmentation with Foundation Models (CVPR 2024 · similarity 0.89)
- Tackling View-Dependent Semantics in 3D Language Gaussian Splatting (ICML 2025 · similarity 0.87)
- Large Spatial Model: End-to-end Unposed Images to Semantic 3D (NeurIPS 2024 · similarity 0.87)
- Cross-Modal and Uncertainty-Aware Agglomeration for Open-Vocabulary 3D Scene Understanding (CVPR 2025 · similarity 0.87)
- EA3D: Online Open-World 3D Object Extraction from Streaming Videos (NeurIPS 2025 · similarity 0.87)
- MESC-3D:Mining Effective Semantic Cues for 3D Reconstruction from a Single Image (CVPR 2025 · 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.
