Research map / 3D vision / 3D generation
WildSeg3D: Segment Any 3D Objects in the Wild from 2D Images
Yansong Guo, Jie Hu, Yansong Qu, Liujuan Cao
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 Dynamic Global Alignment to MASt3R instead of standard GA (ablation) | WildSeg3D (MASt3R) · NVOS | mIoU: 88.6% -> 94.1% | Improved |
| add Dynamic Global Alignment to MASt3R instead of standard GA (ablation) | WildSeg3D (MASt3R) · SPIn-NeRF | mIoU: 89.4% -> 94.0% | Improved |
| add DGA to DUSt3R instead of standard GA (ablation) | DUSt3R-based segmentation · NVOS | mIoU: 83.8% -> 85.7% | Improved |
| add DGA to DUSt3R instead of standard GA (ablation) | DUSt3R-based segmentation · SPIn-NeRF | mIoU: 83.1% -> 84.9% | Improved |
| use DGA instead of sparse global alignment in MASt3R (ablation) | WildSeg3D (MASt3R) · NVOS | mIoU: 85.7% -> 94.1% | Improved |
| use DGA instead of sparse global alignment in MASt3R (ablation) | WildSeg3D (MASt3R) · SPIn-NeRF | mIoU: 86.2% -> 94.0% | Improved |
| use proposed adjustment function instead of Arctan (ablation) | DGA · NVOS | mIoU: 91.3% -> 94.1% | Improved |
| use proposed adjustment function instead of Logarithmic (ablation) | DGA · NVOS | mIoU: 89.5% -> 94.1% | Improved |
| use proposed adjustment function instead of Sigmoid (ablation) | DGA · NVOS | mAcc: 97.8% -> 99.0% | Improved |
| increase input views from 3 to 5 (ablation) | WildSeg3D · NVOS | mIoU: 89.8% -> 94.1% | Improved |
| increase input views from 5 to 9 (ablation) | WildSeg3D · NVOS | mIoU: 94.1% -> 93.8% | No clear effect |
| increase input views from 5 to 7 (ablation) | WildSeg3D · SPIn-NeRF | mIoU: 94.0% -> 94.9% | Improved |
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.
- OmniSeg3D — OmniSeg3D: Omniversal 3D Segmentation via Hierarchical Contrastive Learning (CVPR 2024) · better on 2 of 4 shared results
- SAGA — SAGA: Stochastic Whole-Body Grasping with Contact (ECCV 2022) · better on 4 of 4 shared results
- FlashSplat — FlashSplat: 2D to 3D Gaussian Splatting Segmentation Solved Optimally (ECCV 2024) · better on 2 of 2 shared results
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
- AGILE3D: Attention Guided Interactive Multi-object 3D Segmentation (ICLR 2024 · similarity 0.87)
- WildGaussians: 3D Gaussian Splatting In the Wild (NeurIPS 2024 · similarity 0.86)
- OnlineAnySeg: Online Zero-Shot 3D Segmentation by Visual Foundation Model Guided 2D Mask Merging (CVPR 2025 · similarity 0.86)
- WildCAT3D: Appearance-Aware Multi-View Diffusion in the Wild (NeurIPS 2025 · similarity 0.85)
- Probabilistic Interactive 3D Segmentation with Hierarchical Neural Processes (ICML 2025 · similarity 0.85)
- Mosaic3D: Foundation Dataset and Model for Open-Vocabulary 3D Segmentation (CVPR 2025 · similarity 0.84)
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.
