Research map / Video / Video segmentation and tracking
Language-Bridged Spatial-Temporal Interaction for Referring Video Object Segmentation
Zihan Ding, Tianrui Hui, Junshi Huang, Xiaoming Wei, Jizhong Han, Si Liu
Check 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 language-bridged spatial-temporal interaction in the encoding phase | LBDT-1 · A2D Sentences test | AP: 39.9 -> 46.1 | Improved |
| Use LBDT-4 instead of LBDT-1 (ablation) | LBDT · A2D Sentences test | AP: 46.1 -> 47.2 | Improved |
| Use LBDT-4 instead of LBDT-1 (ablation) | LBDT · J-HMDB Sentences | AP: 40.3 -> 41.1 | Improved |
| Add spatial-to-language-to-temporal transfer (ablation) | Language-bridged duplex transfer · A2D Sentences ablation | AP: 38.8 -> 40.6 | Improved |
| Add temporal-to-language-to-spatial transfer (ablation) | Language-bridged duplex transfer · A2D Sentences ablation | AP: 38.8 -> 43.7 | Improved |
| Add both directions of language-bridged transfer (ablation) | Language-bridged duplex transfer · A2D Sentences ablation | AP: 43.7 -> 44.5 | Improved |
| Add spatial-temporal consistency activation (ablation) | BCA · A2D Sentences ablation | AP: 44.5 -> 45.3 | Improved |
| Add language denoising (ablation) | BCA · A2D Sentences ablation | AP: 44.5 -> 45.0 | Improved |
| Set the frame-difference interval to 6 (ablation) | LBDT-1 · A2D Sentences | AP: 45.0 -> 46.1 | Improved |
| Insert LBDT in stage 4 in addition to stage 5 (ablation) | LBDT · A2D Sentences ablation | AP: 42.3 -> 46.1 | Improved |
| Insert LBDT in stage 3 as well as stages 4 and 5 (ablation) | LBDT · A2D Sentences ablation | AP: 46.1 -> 45.3 | Hurt |
| Insert LBDT in stages 2 and 3 in addition to stages 4 and 5 (ablation) | LBDT · A2D Sentences ablation | AP: 46.1 -> 38.7 | Hurt |
Closest papers by meaning
- Language-Aware Spatial-Temporal Collaboration for Referring Video Segmentation (TPAMI 2023 · similarity 0.94)
- Temporal Collection and Distribution for Referring Video Object Segmentation (ICCV 2023 · similarity 0.91)
- Semantic-Assisted Object Clustering for Multi-Modal Referring Video Segmentation (TPAMI 2025 · similarity 0.90)
- Referring Segmentation via Encoder-Fused Cross-Modal Attention Network (TPAMI 2022 · similarity 0.90)
- SOC: Semantic-Assisted Object Cluster for Referring Video Object Segmentation (NeurIPS 2023 · similarity 0.89)
- Decoupling Static and Hierarchical Motion Perception for Referring Video Segmentation (CVPR 2024 · 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.
