Research map / Detection and segmentation / Open-set and few-shot detection
Segment Everything Everywhere All at Once
Xueyan Zou, Jianwei Yang, Hao Zhang, Feng Li, Linjie Li, Jianfeng Wang, Lijuan Wang, Jianfeng Gao, Yong Jae Lee
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 |
|---|---|---|---|
| SEEM vs SAM (B) | interactive segmentation · PascalVOC | 20-NoC90: 4.12 -> 3.79 | Improved |
| add visual compositional prompt to text | SEEM (T) · RefCOCOg | cIoU: +10.5 cIoU | Improved |
| self-attention composition instead of output ensemble (ablation) | SEEM (T) · RefCOCOg | cIoU: 63.0 -> 71.5 | Improved |
| visual output query instead of text output query (ablation) | SEEM (T) composition · RefCOCOg | cIoU: 66.5 -> 71.5 | Improved |
| add text prompt to visual prompt (ablation) | SEEM (T) · RefCOCOg | cIoU: 70.7 -> 71.5 | Improved |
| train from scratch instead of X-Decoder checkpoint (ablation) | SEEM (T) · RefCOCOg | cIoU: 58.0 -> 51.4 | Hurt |
| train from scratch (ablation) | SEEM (T) · Pascal VOC | NoC90: 4.67 -> 4.41 | Improved |
| increase interactive iterations 1 -> 5 (ablation) | SEEM (T) · Pascal VOC | NoC90: 5.14 -> 4.59 | Improved |
| use LVIS masks (ablation) | SEEM (T) · Pascal VOC | NoC90: 0.3 point gain on NoC@0.9 | Improved |
| more interactive iterations (ablation) | SEEM (T) · COCO | PQ: 50.7 -> 50.6 | No clear effect |
Closest papers by meaning
- Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts (CVPR 2024 · similarity 0.89)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.89)
- Perceive Anything: Recognize, Explain, Caption, and Segment Anything in Images and Videos (NeurIPS 2025 · similarity 0.89)
- Visual In-Context Prompting (CVPR 2024 · similarity 0.89)
- Segment Anyword: Mask Prompt Inversion for Open-Set Grounded Segmentation (ICML 2025 · similarity 0.88)
- Prompt-Driven Referring Image Segmentation with Instance Contrasting (CVPR 2024 · similarity 0.88)
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.
