Research map / Detection and segmentation / CNN features
ReSTR: Convolution-Free Referring Image Segmentation Using Transformers
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · ReferIt test | IoU: 66.70 -> 70.18 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC val | IoU: 65.65 -> 67.22 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC testA | IoU: 68.29 -> 69.30 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC testB | IoU: 63.08 -> 64.45 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC+ val | IoU: 55.50 -> 55.78 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC+ testA | IoU: 59.20 -> 60.44 | Helped |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · UNC+ testB | IoU: 49.36 -> 48.27 | Hurt |
| Use transformer-based visual and language encoders with flexible multimodal fusion | ReSTR · Gref val | IoU: 54.40 -> 54.48 | Helped |
| Indirectly connect visual features and the class seed through linguistic features (ablation) | ReSTR multimodal fusion encoder · Gref val | IoU: 51.27 -> 52.81 | Helped |
| Allow the seed to use visual information via visual-attended linguistic features rather than restricting it to language (ablation) | ReSTR multimodal fusion encoder · Gref val | IoU: 45.89 -> 52.81 | Helped |
| Add the segmentation decoder to refine patch predictions (ablation) | ReSTR with 4 fusion layers · Gref val | IoU: 52.81 -> 54.48 | Helped |
| Add the segmentation decoder to refine patch predictions (ablation) | ReSTR with 2 fusion layers · Gref val | IoU: 48.12 -> 48.43 | Helped |
Closest papers by meaning
- LQMFormer: Language-aware Query Mask Transformer for Referring Image Segmentation (CVPR 2024 · similarity 0.87)
- Mask Grounding for Referring Image Segmentation (CVPR 2024 · similarity 0.87)
- SetCon: Towards Open-Ended Referring Segmentation via Set-Level Concept Prediction (NeurIPS 2026 · similarity 0.87)
- Rethinking Decoders for Transformer-based Semantic Segmentation: A Compression Perspective (NeurIPS 2024 · similarity 0.86)
- Curriculum Point Prompting for Weakly-Supervised Referring Image Segmentation (CVPR 2024 · similarity 0.86)
- Prompt-Driven Referring Image Segmentation with Instance Contrasting (CVPR 2024 · 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.