Research map / Video / Video segmentation and tracking
DVIS++: Improved Decoupled Framework for Universal Video Segmentation
Tao Zhang, Xingye Tian, Yikang Zhou, Shunping Ji, Xuebo Wang, Xin Tao, Yuan Zhang, Pengfei Wan, Zhongyuan Wang, Yu Wu
This paper tried a common change: Add a contrastive loss
add contrastive loss to segmenter · OVIS val (ResNet-50) · AP: 25.0 -> 25.8 No clear effect
What 1,317 accepted papers report for this change
- 1,253 improved
- 22 no clear effect
- 16 mixed
- 26 hurt
- Beyond Looking Up, Try Looking Around: Harmonizing Global Structure and Local Consistency in Optimal Transport for Short Text Clustering ICML 2026 · add contrastive learning · SearchSnippets · ACC: 56.65 -> 80.12Improved
- SleepLM: Natural-Language Intelligence for Human Sleep ICML 2026 · add contrastive objective to captioning (CoCa vs Cap) · sleep event IoU · IoU: 29.2 -> 27.4Hurt
- LIMSSR: LLM-Driven Sequence-to-Score Reasoning under Training-Time Incomplete Multimodal Observations ICML 2026 · add contrastive loss L_con · FS1000 average · Spearman correlation: 0.759 -> 0.789Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 contrastive loss to segmenter (ablation) | MinVIS baseline · OVIS val (ResNet-50) | AP: 25.0 -> 25.8 | No clear effect |
| add referring tracker (ablation) | DVIS++ · OVIS val (ResNet-50) | AP: 25.8 -> 32.8 | Improved |
| add denoising training strategy (ablation) | DVIS++ · OVIS val (ResNet-50) | AP: 32.8 -> 36.5 | Improved |
| add contrastive loss to referring tracker (ablation) | DVIS++ · OVIS val (ResNet-50) | AP: 36.5 -> 37.2 | Improved |
| add temporal refiner (ablation) | DVIS++ · OVIS val (ResNet-50) | AP: 37.2 -> 41.2 | Improved |
| add contrastive loss to temporal refiner (ablation) | DVIS++ · OVIS val (ResNet-50) | AP: 41.2 -> 40.6 | Hurt |
| use referring cross-attention instead of standard (ablation) | referring tracker · OVIS val (ResNet-50) | AP: 28.3 -> 32.8 | Improved |
| match Q_Seg with Hungarian before using as initial value (ablation) | referring tracker · OVIS val (ResNet-50) | AP: 31.9 -> 32.8 | Improved |
| use learnable embedding instead of matched Q_Seg as initial value (ablation) | referring tracker · OVIS val (ResNet-50) | AP: 32.8 -> 33.1 | No clear effect |
| add long-term attention to temporal refiner (ablation) | temporal refiner · OVIS val (ResNet-50) | AP: 37.0 -> 40.2 | Improved |
| add self-attention to temporal refiner (ablation) | temporal refiner · OVIS val (ResNet-50) | AP: 38.5 -> 40.2 | Improved |
| add cross-attention to temporal refiner (ablation) | temporal refiner · OVIS val (ResNet-50) | AP: 39.4 -> 40.2 | 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.
- DVIS — DVIS: Decoupled Video Instance Segmentation Framework (ICCV 2023) · better on 3 of 3 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
- VidSeg: Training-free Video Semantic Segmentation based on Diffusion Models (CVPR 2025 · similarity 0.88)
- Decoupled Motion Expression Video Segmentation (CVPR 2025 · similarity 0.88)
- SyncVIS: Synchronized Video Instance Segmentation (NeurIPS 2024 · similarity 0.88)
- Minimizing Labeled, Maximizing Unlabeled: An Image-Driven Approach for Video Instance Segmentation (CVPR 2025 · similarity 0.87)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 2025 · similarity 0.87)
- OpenVIS: Open-vocabulary Video Instance 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.
