Research map / Video / Video segmentation and tracking
Modular Interactive Video Object Segmentation: Interaction-to-Mask, Propagation and Difference-Aware Fusion
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 top-k memory filtering (ablation) | MiVOS propagation · DAVIS 2017 multi-object semi-supervised validation | J&F: 81.5 -> 83.1 | Helped |
| Add BL30K training (ablation) | MiVOS propagation · DAVIS 2017 multi-object semi-supervised validation | J&F: 81.5 -> 83.8 | Helped |
| Add both top-k filtering and BL30K training (ablation) | MiVOS propagation · DAVIS 2017 multi-object semi-supervised validation | J&F: 81.5 -> 84.5 | Helped |
| Add top-k memory filtering (ablation) | MiVOS interactive baseline · DAVIS interactive validation | AUC-J&F: 86.0 -> 87.2 | Helped |
| Add BL30K pretraining (ablation) | MiVOS interactive baseline · DAVIS interactive validation | AUC-J&F: 86.0 -> 87.4 | Helped |
| Add learnable fusion (ablation) | MiVOS interactive baseline · DAVIS interactive validation | AUC-J&F: 86.0 -> 87.6 | Helped |
| Add difference-aware guidance to learnable fusion (ablation) | MiVOS interactive model · DAVIS interactive validation | AUC-J&F: 87.6 -> 87.9 | Helped |
| Replace the interaction module with ground-truth masks (ablation) | MiVOS · DAVIS interactive validation | AUC-J&F: 87.9 -> 90.2 | Helped |
| Use the full modular MiVOS method rather than the strongest reported baseline | DAVIS interactive track · DAVIS interactive validation | AUC-J&F: 80.9 -> 87.9 | Helped |
| Use MiVOS-S2M instead of ATNet | Interactive video segmentation · DAVIS 2017 multi-object validation user study | Sum of incremental IoU improvements: 75.1 -> 86.9 | Helped |
Closest papers by meaning
- Referred by Multi-Modality: A Unified Temporal Transformer for Video Object Segmentation (AAAI 2024 · similarity 0.87)
- Dual Prototype Attention for Unsupervised Video Object Segmentation (CVPR 2024 · similarity 0.87)
- M^3-VOS: Multi-Phase, Multi-Transition, and Multi-Scenery Video Object Segmentation (CVPR 2025 · similarity 0.87)
- Point-VOS: Pointing Up Video Object Segmentation (CVPR 2024 · similarity 0.86)
- RMem: Restricted Memory Banks Improve Video Object Segmentation (CVPR 2024 · similarity 0.86)
- Multi-Granularity Video Object Segmentation (AAAI 2025 · 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.