Research map / 3D vision / 3D generation
Online 3D Instance Segmentation at task-oriented granularity with Unposed Monocular Video
Dong Wu, Baicheng Li, Yingdian Cao, Shunkai Zhou, Yiwen Lu, Hongbin Zha
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 geometric association metric (GAM) (ablation) | Ours · Replica | AP50: 17.3 -> 23.4 | Improved |
| add semantic similarity metric (SSM) (ablation) | Ours · Replica | AP50: 21.7 -> 23.4 | Improved |
| add mutual exclusivity metric (MEM) (ablation) | Ours · Replica | AP50: 8.2 -> 23.4 | Improved |
| add MEM (ablation) | Ours · Replica | AP25: 12.0 -> 37.3 | Improved |
| add IoU-based cluster merging (ablation) | Ours · Replica | AP50: 13.8 -> 23.4 | Improved |
| add priority-ordered mask merging (descending confidence) (ablation) | Ours · Replica | AP50: 20.3 -> 23.4 | Improved |
| add priority-ordered mask merging (ablation) | Ours · Replica | AP25: 33.8 -> 37.3 | Improved |
| replace GT pose/depth with MASt3R-SLAM input for OnlineAnySeg | Ours · ScanNet200 val (class-agnostic) | AP50: 36.1 -> 4.7 | Hurt |
| replace GT pose/depth with MASt3R-SLAM input for Open-YOLO 3D | Ours · ScanNet200 val (open-vocabulary) | AP50: 31.7 -> 0.8 | Hurt |
| use our online unposed method instead of offline PanSt3R | Ours · Replica | AP50: 20.9 -> 23.4 | Improved |
Closest papers by meaning
- Any3DIS: Class-Agnostic 3D Instance Segmentation by 2D Mask Tracking (CVPR 2025 · similarity 0.88)
- EmbodiedSAM: Online Segment Any 3D Thing in Real Time (ICLR 2025 · similarity 0.88)
- Online Segment Any 3D Thing as Instance Tracking (NeurIPS 2025 · similarity 0.88)
- Open3DIS: Open-Vocabulary 3D Instance Segmentation with 2D Mask Guidance (CVPR 2024 · similarity 0.87)
- A Unified Framework for 3D Scene Understanding (NeurIPS 2024 · similarity 0.87)
- ODIN: A Single Model for 2D and 3D Segmentation (CVPR 2024 · similarity 0.86)
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.
