Research map / Video / Video segmentation and tracking
Efficient Regional Memory Network for Video Object Segmentation
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 |
|---|---|---|---|
| use local-to-local regional matching in memory and query instead of global-to-global matching (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.818 -> 0.835 | Helped |
| use local-to-local regional matching in memory and query instead of global-to-global matching (ablation) | RMNet · DAVIS 2017 val | Time (ms): 10.68 -> 2.09 | Helped |
| restrict matching to regional memory features (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.818 -> 0.822 | Helped |
| restrict matching to the predicted query region (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.818 -> 0.828 | Helped |
| predict the query region with optical flow rather than the previous-frame region (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.792 -> 0.835 | Helped |
| predict the query region with optical flow rather than best-match pixels (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.819 -> 0.835 | Helped |
| replace TinyFlowNet with FlowNet2-CSS (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.835 -> 0.837 | No clear effect |
| replace TinyFlowNet with RAFT (ablation) | RMNet · DAVIS 2017 val | Avg.: 0.835 -> 0.834 | No clear effect |
| use TinyFlowNet rather than FlowNet2-CSS for region prediction (ablation) | RMNet · DAVIS 2017 val | Time (ms): 59.93 -> 10.05 | Helped |
| use TinyFlowNet rather than RAFT for region prediction (ablation) | RMNet · DAVIS 2017 val | Time (ms): 157.78 -> 10.05 | Helped |
Closest papers by meaning
- Learning Position and Target Consistency for Memory-Based Video Object Segmentation (CVPR 2021 · similarity 0.90)
- LiVOS: Light Video Object Segmentation with Gated Linear Matching (CVPR 2025 · similarity 0.89)
- SwiftNet: Real-Time Video Object Segmentation (CVPR 2021 · similarity 0.89)
- Rethinking Space-Time Networks with Improved Memory Coverage for Efficient Video Object Segmentation (NeurIPS 2021 · similarity 0.89)
- RMem: Restricted Memory Banks Improve Video Object Segmentation (CVPR 2024 · similarity 0.87)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 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.