Research map / Video / Video segmentation and tracking
OnlineRefer: A Simple Online Baseline for Referring Video Object Segmentation
Dongming Wu, Tiancai Wang, Yuang Zhang, Xiangyu Zhang, Jianbing Shen
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 query update (content query from previous output embedding) (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 49.8 -> 57.3 | Improved |
| also update position embedding across frames (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 57.3 -> 32.1 | Hurt |
| add cross-frame query propagation (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 49.3 -> 57.3 | Improved |
| propagate one selected query instead of concatenating with initial queries (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 55.2 -> 57.3 | Improved |
| propagate top-1 query instead of top-4 (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 55.2 -> 57.3 | Improved |
| increase initial queries from 1 to 5 (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 54.1 -> 57.3 | Improved |
| increase initial queries from 5 to 8 (ablation) | OnlineRefer · Refer-Youtube-VOS | J&F: 57.3 -> 56.9 | No clear effect |
| online query propagation instead of offline clip-level query sharing | ReferFormer (ResNet-50) · Refer-Youtube-VOS | J&F: 55.6 -> 57.3 | Improved |
| online query propagation instead of offline (Swin-L) on Refer-DAVIS17 | ReferFormer (Swin-L) · Refer-DAVIS17 | J&F: 60.5 -> 64.8 | Improved |
| semi-online query propagation across clips with Video-Swin-B | ReferFormer (Video-Swin-B) · Refer-Youtube-VOS | J&F: 62.9 -> 62.9 | No clear effect |
Later papers that compared with this method
3 later accepted papers report OnlineRefer as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- AR2-4FV: Anchored Referring and Re-identification for Long-Term Grounding in Fixed-View Videos (CVPR 2026 · better on 2 of 2 shared results)
- One Token to Seg Them All: Language Instructed Reasoning Segmentation in Videos (NeurIPS 2024 · better on 2 of 2 shared results)
- VISA: Reasoning Video Object Segmentation via Large Language Model (ECCV 2024 · better on 0 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
- LoSh: Long-Short Text Joint Prediction Network for Referring Video Object Segmentation (CVPR 2024 · similarity 0.93)
- Long-RVOS: A Comprehensive Benchmark for Long-term Referring Video Object Segmentation (CVPR 2026 · similarity 0.93)
- End-to-End Referring Video Object Segmentation With Multimodal Transformers (CVPR 2022 · similarity 0.92)
- Learning Cross-Modal Affinity for Referring Video Object Segmentation Targeting Limited Samples (ICCV 2023 · similarity 0.92)
- SAMWISE: Infusing Wisdom in SAM2 for Text-Driven Video Segmentation (CVPR 2025 · similarity 0.90)
- Semantic and Sequential Alignment for Referring Video Object Segmentation (CVPR 2025 · similarity 0.90)
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.
