Research map / Detection and segmentation / Open-set and few-shot detection
InstMove: Instance Motion for Object-Centric Video Segmentation
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 instance-level motion scoring to CrossVIS tracking | CrossVIS · OVIS validation | AP: 12.6 -> 16.7 | Helped |
| Add instance-level motion scoring to MinVIS tracking | MinVIS · OVIS validation | AP: 26.2 -> 27.6 | Helped |
| Add instance-level motion scoring to IDOL tracking | IDOL · OVIS validation | AP: 29.2 -> 30.7 | Helped |
| Add instance-level motion scoring to MinVIS tracking | MinVIS · YouTube-VIS-Long validation | AP: 22.3 -> 25.9 | Helped |
| Add instance-level motion scoring to IDOL tracking | IDOL · YouTube-VIS-Long validation | AP: 35.7 -> 40.6 | Helped |
| Add instance-level motion scoring to Unicorn tracking | Unicorn · BDD100K validation | ID switch: 2044 -> 1460 | Helped |
| Add instance-level motion masks to the segmentation decoder | STCN · DAVIS 2017 validation | J&F-Mean: 83.8 -> 85.1 | Helped |
| Use instance-mask motion prediction instead of optical-flow mask propagation | Motion prediction · OVIS-Sub motion prediction | mAP: 46.3 -> 67.8 | Helped |
| Use instance-mask motion prediction instead of optical-flow mask propagation | Motion prediction · OVIS-Sub-Sparse motion prediction | mAP: 30.6 -> 57.3 | Helped |
| Add image encoder features to refine predicted mask boundaries (ablation) | InstMove · OVIS-Sub motion prediction | mAP: 61.0 -> 67.8 | Helped |
| Add image encoder features to refine predicted mask boundaries (ablation) | InstMove · OVIS-Sub-Sparse motion prediction | mAP: 46.6 -> 57.3 | Helped |
| Use fixed pretrained image encoder features (ablation) | IDOL+InstMove · OVIS validation | AP: 29.2 -> 30.9 | Helped |
Closest papers by meaning
- InsPro: Propagating Instance Query and Proposal for Online Video Instance Segmentation (NeurIPS 2022 · similarity 0.90)
- OpenVIS: Open-vocabulary Video Instance Segmentation (AAAI 2025 · similarity 0.87)
- TrackIME: Enhanced Video Point Tracking via Instance Motion Estimation (NeurIPS 2024 · similarity 0.87)
- Decoupled Motion Expression Video Segmentation (CVPR 2025 · similarity 0.86)
- Segment Any Motion in Videos (CVPR 2025 · similarity 0.86)
- GMOS: Grounding Moving Object Segmentation in 3D Space and Time (NeurIPS 2026 · 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.