Research map / Video / Video segmentation and tracking
Video Mask Transfiner for High-Quality Video Instance Segmentation
Lei Ke, Henghui Ding, Martin Danelljan, Yu-Wing Tai, Chi-Keung Tang, Fisher Yu
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 |
|---|---|---|---|
| use quadtree sequence grouping across frames (clip length 5) (ablation) | VMT · HQ-YTVIS val | AP^B: 30.2 -> 31.4 | Improved |
| use quadtree sequence grouping across frames (all frames) (ablation) | VMT · HQ-YTVIS val | AP^B: 32.3 -> 33.7 | Improved |
| lengthen the clip from 1 frame to all frames (ablation) | VMT · HQ-YTVIS val | AP^B: 26.1 -> 33.7 | Improved |
| use video-level instead of frame-level instance queries in the detector (ablation) | VMT · HQ-YTVIS val | AP^B: 31.8 -> 33.7 | Improved |
| use video-level instance query instead of fixed FCN detector (ablation) | VMT · HQ-YTVIS val | AP^B: 32.2 -> 33.7 | Improved |
| refine 3D incoherent regions instead of detected object boundaries (ablation) | VMT · HQ-YTVIS val | AP^B: 32.8 -> 33.7 | Improved |
| train on iteratively self-corrected HQ-YTVIS instead of YTVIS | VMT · HQ-YTVIS val | AP^B: 30.5 -> 33.7 | Improved |
| train on HQ-YTVIS instead of YTVIS | SeqFormer · HQ-YTVIS val | AP^B: 28.9 -> 31.0 | Improved |
| train on HQ-YTVIS instead of YTVIS | MaskTrack · HQ-YTVIS val | AP^B: 19.8 -> 21.7 | Improved |
| train on HQ-YTVIS instead of YTVIS | VMT · OVIS val | AP^M: 15.9 -> 17.1 | Improved |
| iterative self-correction with VMT instead of MaskTrack or SeqFormer (ablation) | VMT · HQ-YTVIS val | AP^B: around 60.0 for baselines | Improved |
| replace SeqFormer with VMT on OVIS (Swin-L) | SeqFormer · OVIS val | AP: 17.9 -> 19.8 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- SeqFormer — SeqFormer: Sequential Transformer for Video Instance Segmentation (ECCV 2022) · better on 12 of 12 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
- Mask-Free Video Instance Segmentation (CVPR 2023 · similarity 0.94)
- InstanceFormer: An Online Video Instance Segmentation Framework (AAAI 2023 · similarity 0.91)
- End-to-End Video Instance Segmentation With Transformers (CVPR 2021 · similarity 0.91)
- Video Instance Segmentation (ICCV 2019 · similarity 0.90)
- Classifying, Segmenting, and Tracking Object Instances in Video with Mask Propagation (CVPR 2020 · similarity 0.90)
- Robust and Consistent Online Video Instance Segmentation via Instance Mask Propagation (AAAI 2025 · similarity 0.89)
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.
