Research map / Video / Video segmentation and tracking
OV-VIS: Open-Vocabulary Video Instance Segmentation
Haochen Wang, Cilin Yan, Keyan Chen, Xiaolong Jiang, Xu Tang, Yao Guang Hu, Guoliang Kang, Weidi Xie, Efstratios Gavves
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 InstCLIP (ablation) | MinVIS-CLIP · BURST | AP: 2.1 -> 3.3 | Improved |
| add InstCLIP (ablation) | MinVIS-CLIP · UVO | AP: 9.0 -> 13.2 | Improved |
| add InstCLIP (ablation) | MinVIS-CLIP · YouTube-VIS | AP: 30.6 -> 48.6 | Improved |
| add contrastive instance margin loss L_SC (ablation) | InstFormer · BURST | AP: 3.3 -> 3.5 | Improved |
| add InstCLIP token (ablation) | InstFormer · BURST | AP: 3.5 -> 3.9 | Improved |
| add rollout tracker (ablation) | InstFormer · BURST | AP: 3.9 -> 4.2 | Improved |
| add instance tokens without guidance to CLIP (ablation) | N times CLIP · BURST | AP: 2.11 -> 1.09 | Hurt |
| add designed instance attention mask guidance (ablation) | InstCLIP · BURST | AP: 1.70 -> 3.28 | Improved |
| add register token (ablation) | InstCLIP · BURST | AP: 3.28 -> 3.87 | Improved |
| add instance token to rollout tracker (ablation) | Rollout tracker · BURST | AP: 3.6 -> 4.2 | Improved |
| use RNN instead of linear layer in rollout tracker (ablation) | Rollout tracker · BURST | AP: 3.7 -> 4.2 | Improved |
| use RNN instead of two-layer MLP in rollout tracker (ablation) | Rollout tracker · BURST | AP: 3.4 -> 4.2 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- MinVIS — MinVIS: A Minimal Video Instance Segmentation Framework without Video-based Training (NeurIPS 2022) · better on 2 of 2 shared results
- DVIS — DVIS: Decoupled Video Instance Segmentation Framework (ICCV 2023) · better on 1 of 1 shared result
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
- OpenVIS: Open-vocabulary Video Instance Segmentation (AAAI 2025 · similarity 0.99)
- Towards Open-Vocabulary Video Instance Segmentation (ICCV 2023 · similarity 0.94)
- Unified Embedding Alignment for Open-Vocabulary Video Instance Segmentation (ECCV 2024 · similarity 0.92)
- OW-VISCapTor: Abstractors for Open-World Video Instance Segmentation and Captioning (NeurIPS 2024 · similarity 0.89)
- Transferable and Principled Efficiency for Open-Vocabulary Segmentation (CVPR 2024 · similarity 0.88)
- OVTR: End-to-End Open-Vocabulary Multiple Object Tracking with Transformer (ICLR 2025 · similarity 0.87)
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.
