Research map / Detection and segmentation / Open-set and few-shot detection
Deep Interactive Object Selection
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 the interactive FCN-8s method instead of the strongest listed baseline | interactive object selection · Pascal VOC 2012 validation (85% IU) | mean number of clicks: 11.37 -> 6.88 | Helped |
| use the interactive FCN-8s method instead of the strongest listed baseline | interactive object selection · Grabcut (90% IU) | mean number of clicks: 8.38 -> 6.04 | Helped |
| use the interactive FCN-8s method instead of the strongest listed baseline | interactive object selection · Berkeley (90% IU) | mean number of clicks: 14.02 -> 8.65 | Helped |
| use the interactive FCN-8s method instead of the strongest listed baseline | interactive object selection · MS COCO seen categories (85% IU) | mean number of clicks: 13.90 -> 8.31 | Helped |
| use the interactive FCN-8s method instead of the strongest listed baseline | interactive object selection · MS COCO unseen categories (85% IU) | mean number of clicks: 14.86 -> 7.82 | Helped |
| combine user clicks with an interactive FCN instead of applying semantic segmentation directly | MS COCO seen categories · MS COCO seen categories (single positive click) | mean IU accuracy: 47.01% -> 48.35% | Helped |
| combine user clicks with an interactive FCN instead of applying semantic segmentation directly | MS COCO unseen categories · MS COCO unseen categories (single positive click) | mean IU accuracy: 16.14% -> 42.94% | Helped |
| replace FCN-32s with FCN-16s (ablation) | interactive FCN · interactive segmentation | segmentation accuracy | Helped |
| replace FCN-16s with FCN-8s (ablation) | interactive FCN · interactive segmentation | segmentation accuracy | No clear effect |
| train a finer-stride FCN than FCN-8s (ablation) | interactive FCN · interactive segmentation | segmentation accuracy | No clear effect |
| initialize the two added input-channel filters with their mean value instead of zeros (ablation) | interactive FCN · interactive segmentation | segmentation accuracy | No clear effect |
| combine all three negative-click sampling strategies instead of relying on any one (ablation) | interactive FCN training · interactive segmentation | segmentation accuracy | Helped |
Closest papers by meaning
- Fast User-Guided Video Object Segmentation by Interaction-And-Propagation Networks (CVPR 2019 · similarity 0.87)
- Order-aware Interactive Segmentation (ICLR 2025 · similarity 0.85)
- TETRIS: Towards Exploring the Robustness of Interactive Segmentation (AAAI 2024 · similarity 0.84)
- NTClick: Achieving Precise Interactive Segmentation With Noise-tolerant Clicks (CVPR 2025 · similarity 0.83)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.82)
- Interacted Object Grounding in Spatio-Temporal Human-Object Interactions (AAAI 2025 · similarity 0.82)
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.