Research map / Detection and segmentation / Open-set and few-shot detection
FocalClick: Towards Practical Interactive Image Segmentation
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 FocalClick with SegFormer-B3-S2 instead of RITM-hrnet32 | Interactive segmentation · DAVIS | NoC 85: 4.11 -> 3.61 | Helped |
| Use FocalClick with SegFormer-B3-S2 instead of RITM-hrnet32 | Interactive mask correction · DAVIS-585, from initial mask | NoC85: 3.68 -> 2.00 | Helped |
| Use FocalClick with SegFormer-B3-S2 instead of RITM-hrnet32 | Interactive segmentation · GrabCut | NoC 85: 1.46 -> 1.44 | Helped |
| Use localized Segmentor inference with FocalClick-B0-S1 rather than HRNet-18s-400 full-image inference | CPU laptop segmentation · CPU laptop | Speed/ms: 470 -> 41 | Helped |
| Add Target Crop (ablation) | Naive-B0-S1 · DAVIS | NoC85: 10.70 -> 5.70 | Helped |
| Add Focus Crop after Target Crop (ablation) | Naive-B0-S1 with Target Crop · DAVIS | NoC85: 5.70 -> 5.15 | Helped |
| Add Progressive Merge after Target Crop and Focus Crop (ablation) | Naive-B0-S1 with TC and FC · DAVIS-585 | NoC85: 5.41 -> 2.63 | Helped |
| Add Target Crop (ablation) | Naive-B0-S2 · DAVIS | NoC85: 5.24 -> 4.52 | Helped |
| Add Focus Crop after Target Crop (ablation) | Naive-B0-S2 with Target Crop · DAVIS | NoC85: 4.52 -> 4.15 | Helped |
| Add Progressive Merge after Target Crop and Focus Crop (ablation) | Naive-B0-S2 with TC and FC · DAVIS-585 | NoC85: 3.94 -> 2.21 | Helped |
| Use small S1 input without cropping modules rather than S2 input (ablation) | Naive-B0 · DAVIS | NoC85: 5.24 -> 10.70 | Hurt |
| Remove Target Crop and Focus Crop (ablation) | FocalClick-B0-S1 · DAVIS | NoC80/90: 5.13/7.42 -> 10.70/15.60 | Hurt |
Closest papers by meaning
- NTClick: Achieving Precise Interactive Segmentation With Noise-tolerant Clicks (CVPR 2025 · similarity 0.90)
- Variance-Insensitive and Target-Preserving Mask Refinement for Interactive Image Segmentation (AAAI 2024 · similarity 0.88)
- FocSAM: Delving Deeply into Focused Objects in Segmenting Anything (CVPR 2024 · similarity 0.86)
- Click2Mask: Local Editing with Dynamic Mask Generation (AAAI 2025 · similarity 0.86)
- CFR-ICL: Cascade-Forward Refinement with Iterative Click Loss for Interactive Image Segmentation (AAAI 2024 · similarity 0.86)
- Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts (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.