Research map / Video / Video segmentation and tracking
VP-SAM: Taming Segment Anything Model for Video Polyp Segmentation via Disentanglement and Spatio-temporal Side Network
Zhixue Fang, Yuzhi Liu, Huisi Wu, Jing Qin
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 |
|---|---|---|---|
| add semantic disentanglement adapter (ESE and SDI together) (ablation) | VP-SAM · SUN-SEG-Easy | mDice: 85.76 -> 87.56 | Improved |
| add semantic disentanglement adapter (ESE and SDI together) (ablation) | VP-SAM · SUN-SEG-Hard | mDice: 85.68 -> 87.04 | Improved |
| add spatio-temporal side network (DRM plus STI) (ablation) | VP-SAM · SUN-SEG-Easy | mDice: 82.61 -> 87.56 | Improved |
| use DRM and STI together rather than individually (ablation) | VP-SAM STSN · SUN-SEG-Easy | mDice: 85.87 -> 87.56 | Improved |
| increase clip length from 1 to 3 frames | VP-SAM · SUN-SEG | mDice | Improved |
| increase clip length to 5 or 7 | VP-SAM · SUN-SEG | mDice | Hurt |
| use more point prompts | VP-SAM · SUN-SEG | mDice | No clear effect |
| use proposed STSN instead of ST-Adapter (ablation) | SAM · SUN-SEG-Easy | mDice: 84.17 -> 87.56 | Improved |
| use point/mask trackers to extend SAM to video (ablation) | SAM · SUN-SEG-Easy | mDice: 58.85 and 60.31 vs 87.56 | Hurt |
| add spatio-temporal modeling overhead vs lighter SAM adapters | VP-SAM · SUN-SEG-Hard | FPS: 53.63 (SAMed) -> 38.23 | Hurt |
Later papers that compared with this method
2 later accepted papers report VP-SAM as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- WavePolyp: Video Polyp Segmentation via Hierarchical Wavelet-Based Feature Aggregation and Inter-Frame Divergence Perception (ICLR 2026 · better on 6 of 6 shared results)
- FSD-Net: Foundation-Guided Spatiotemporal Distillation for Video Polyp Segmentation (ECCV 2026 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- FST-SAM3: Taming SAM~3 with Frequency-Spatio-Temporal Refinement for Video Polyp Segmentation (ECCV 2026 · similarity 0.91)
- An Embedding-Unleashing Video Polyp Segmentation Framework via Region Linking and Scale Alignment (AAAI 2024 · similarity 0.90)
- Boost the Inference with Co-training: A Depth-guided Mutual Learning Framework for Semi-supervised Medical Polyp Segmentation (CVPR 2025 · similarity 0.84)
- SAM 2: Segment Anything in Images and Videos (ICLR 2025 · similarity 0.83)
- Polyper: Boundary Sensitive Polyp Segmentation (AAAI 2024 · similarity 0.83)
- SAM-I2V: Upgrading SAM to Support Promptable Video Segmentation with Less than 0.2% Training Cost (CVPR 2025 · similarity 0.83)
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.
