Research map / Video / Video segmentation and tracking
Continuous Copy-Paste for One-Stage Multi-Object Tracking and Segmentation
Zhenbo Xu, Ajin Meng, Zhenbo Shi, Wei Yang, Zhi Chen, Liusheng Huang
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 |
|---|---|---|---|
| train with continuous copy-paste (CCP) | TRCNN · KITTI MOTS leaderboard (cars) | sMOTSA: 67.00 -> 72.05 | Improved |
| train with continuous copy-paste (CCP) | TRCNN · KITTI MOTS leaderboard (pedestrians) | sMOTSA: 47.30 -> 48.37 | Improved |
| train with continuous copy-paste (CCP) | PointTrack · KITTI MOTS leaderboard (cars) | sMOTSA: 78.50 -> 81.52 | Improved |
| train with continuous copy-paste (CCP) | PointTrack · KITTI MOTS leaderboard (pedestrians) | sMOTSA: 61.50 -> 64.34 | Improved |
| add reconstruction decoder (ablation) | CCPNet · KITTI MOTS validation (pedestrians) | MOTSA: 84.55 -> 85.69 | Improved |
| add reconstruction decoder (ablation) | CCPNet · KITTI MOTS validation (pedestrians) | sMOTSA: 68.35 -> 69.35 | Improved |
| add self-training | CCPNet · KITTI MOTS leaderboard (pedestrians) | sMOTSA: 70.16 -> 70.55 | Improved |
| add self-training | CCPNet · KITTI MOTS leaderboard (cars) | sMOTSA: 84.47 -> 84.47 | No clear effect |
| add boundary shift (ablation) | CCPNet · KITTI MOTS validation (pedestrians) | sMOTSA: 68.65 -> 68.76 | Improved |
| use n=3 frames instead of n=2 (ablation) | CCPNet · KITTI MOTS validation (cars) | sMOTSA: 87.33 -> 87.36 | No clear effect |
| raise max instance blocks from 5 to 10 (ablation) | CCPNet · KITTI MOTS validation (pedestrians) | sMOTSA: 67.98 -> 68.39 | Improved |
| raise max instance blocks from 15 to 25 (ablation) | CCPNet · KITTI MOTS validation (pedestrians) | sMOTSA: 68.76 -> 67.49 | Hurt |
Closest papers by meaning
- MOTS: Multi-Object Tracking and Segmentation (CVPR 2019 · similarity 0.90)
- Learning Multi-Object Tracking and Segmentation From Automatic Annotations (CVPR 2020 · similarity 0.90)
- Segment as Points for Efficient Online Multi-Object Tracking and Segmentation (ECCV 2020 · similarity 0.89)
- Simple Copy-Paste Is a Strong Data Augmentation Method for Instance Segmentation (CVPR 2021 · similarity 0.88)
- Segment as Points for Efficient and Effective Online Multi-Object Tracking and Segmentation (TPAMI 2021 · similarity 0.88)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · similarity 0.85)
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.
