Research map / Detection and segmentation / Open-set and few-shot detection
Segment Any Events with Language
Seungjun Lee, Gim Hee Lee
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 text-domain guidance to visual-only guidance (ablation) | SEAL · DSEC11-Ins (box) | AP: 24.0 -> 28.8 | Improved |
| add visual-domain guidance to text-only guidance (ablation) | SEAL · DDD17-Ins (box) | AP: 35.2 -> 38.2 | Improved |
| add Spatial Encoding and Mask Feature Enhancer to fusion-only model (ablation) | SEAL · DSEC-Part (box) | AP: 14.9 -> 18.3 | Improved |
| add Spatial Encoding to fusion-only model (ablation) | SEAL · DSEC11-Ins (box) | AP: 24.3 -> 26.2 | Improved |
| add Mask Feature Enhancer to fusion-only model (ablation) | SEAL · DSEC11-Ins (box) | AP: 24.3 -> 27.7 | Improved |
| add backbone language fusion to SE+MFE model (ablation) | SEAL · DSEC11-Ins (box) | AP: 27.1 -> 28.8 | Improved |
| add part-level guidance to instance+semantic guidance (ablation) | SEAL · DSEC-Part (box) | AP: 15.4 -> 18.3 | Improved |
| add semantic-level guidance to part+instance guidance (ablation) | SEAL · DSEC11-Ins (point) | AP: 21.5 -> 22.4 | Improved |
| add semantic-level guidance to part+instance guidance (ablation) | SEAL · DSEC-Part (box) | AP: 18.5 -> 18.3 | No clear effect |
| add instance-level guidance to semantic-only guidance (ablation) | SEAL · DSEC11-Ins (box) | AP: 28.2 -> 28.3 | Mixed |
| use SEAL single-backbone design instead of frame2voxel | frame2voxel · Efficiency | inference time (ms): 88.19 -> 22.28 | Improved |
Closest papers by meaning
- OpenESS: Event-based Semantic Scene Understanding with Open Vocabularies (CVPR 2024 · similarity 0.91)
- OpenWorldSAM: Extending SAM2 for Universal Image Segmentation with Language Prompts (NeurIPS 2025 · similarity 0.87)
- Talk2Event: Grounded Understanding of Dynamic Scenes from Event Cameras (NeurIPS 2025 · similarity 0.87)
- Segment Any 3D Object with Language (ICLR 2025 · similarity 0.87)
- ESEG: Event-Based Segmentation Boosted by Explicit Edge-Semantic Guidance (AAAI 2025 · similarity 0.87)
- OpenVIS: Open-vocabulary Video Instance Segmentation (AAAI 2025 · 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.
