Research map / Video / Video segmentation and tracking
Weakly Supervised Instance Segmentation for Videos With Temporal Mask Consistency
Check 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 flow-amplified CAMs and flow-boundary loss (ablation) | IRN · YTVIS training data | AP50: 25.42 -> 28.45 | Helped |
| add flow-amplified CAMs and flow-boundary loss (ablation) | IRN · Cityscapes training data | AP50: 8.46 -> 10.75 | Helped |
| add flow-boundary loss (ablation) | IRN · YTVIS training data | AP50: 25.42 -> 26.60 | Helped |
| add flow-amplified CAMs (ablation) | IRN · YTVIS training data | AP50: 25.42 -> 27.47 | Helped |
| add flow-boundary loss (ablation) | IRN · Cityscapes training data | AP50: 8.46 -> 9.51 | Helped |
| add flow-amplified CAMs (ablation) | IRN · Cityscapes training data | AP50: 8.46 -> 10.55 | Helped |
| add intra-frame matching (ablation) | MaskConsist · YTVIS | AP50: 31.08 -> 34.66; 33.75 -> 34.66 | Mixed |
| add inter-frame matching (ablation) | MaskConsist · YTVIS | AP50: 31.08 -> 34.66 | Helped |
| add IoM-NMS (ablation) | MaskConsist · YTVIS | AP50: 33.65 -> 34.66 | Helped |
| add intra-frame matching (ablation) | MaskConsist · Cityscapes | AP50: 14.43 -> 16.05 | Helped |
| add inter-frame matching (ablation) | MaskConsist · Cityscapes | AP50: 14.43 -> 16.05 | Helped |
| add IoM-NMS (ablation) | MaskConsist · Cityscapes | AP50: 15.27 -> 16.05 | Helped |
Closest papers by meaning
- Learning to Track Instances without Video Annotations (CVPR 2021 · similarity 0.91)
- Mask-Free Video Instance Segmentation (CVPR 2023 · similarity 0.90)
- The Devil Is in the Points: Weakly Semi-Supervised Instance Segmentation via Point-Guided Mask Representation (CVPR 2023 · similarity 0.90)
- Instance-Level Video Segmentation From Object Tracks (CVPR 2016 · similarity 0.90)
- Minimizing Labeled, Maximizing Unlabeled: An Image-Driven Approach for Video Instance Segmentation (CVPR 2025 · similarity 0.89)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.88)
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.