Research map / Detection and segmentation / Weakly supervised detection and segmentation
Multi-granularity Interaction Simulation for Unsupervised Interactive Segmentation
Kehan Li, Yian Zhao, Zhennan Wang, Zesen Cheng, Peng Jin, Xiangyang Ji, Li Yuan, Chang Liu, Jie Chen
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 multi-granularity proposals (decay coefficient 0.9 versus 0.0) (ablation) | MIS · DAVIS | NoC@90: 9.92 -> 8.44 | Improved |
| add multi-granularity proposals (alpha 0.9 versus 0.0) (ablation) | MIS · SBD | NoC@90: 10.87 -> 9.51 | Improved |
| add smoothness constraint (ablation) | MIS · DAVIS | NoC@90: 10.08 -> 8.44 | Improved |
| add smoothness constraint (ablation) | MIS · Berkeley | NoC@90: 5.51 -> 4.58 | Improved |
| use top-down sampling instead of uniform sampling of tree nodes (ablation) | MIS · GrabCut | NoC@90: 4.32 -> 2.32 | Improved |
| use top-down sampling instead of uniform sampling (ablation) | MIS · DAVIS | NoC@90: 10.10 -> 8.44 | Improved |
| add connectivity constraint in bottom-up merging (ablation) | MIS · preprocessing speed | images/s: 1.10 -> 3.73 | Improved |
| add connectivity constraint in bottom-up merging (ablation) | MIS · SBD / Berkeley | NoC@90: 9.46 -> 9.51 (SBD NoC@90) | No clear effect |
| add connectivity constraint in bottom-up merging (ablation) | MIS · DAVIS | NoC@90: 7.56 -> 8.44 | Hurt |
| increase decay coefficient to 0.95 (ablation) | MIS · GrabCut | NoC@90: 2.32 -> 2.62 | Hurt |
| increase smoothness weight lambda to 15.0 (ablation) | MIS · DAVIS | NoC@90: 8.44 -> 9.10 | Hurt |
| use MIS pretraining instead of MAE init with 1% labels | SimpleClick-style model · SBD (1% labels) | NoC@90: 10.43 -> 8.15 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TokenCut — TokenCut: Segmenting Objects in Images and Videos With Self-Supervised Transformer and Normalized Cut (TPAMI 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Unified Granularity Controller for Interactive Segmentation (TPAMI 2025 · similarity 0.89)
- GraCo: Granularity-Controllable Interactive Segmentation (CVPR 2024 · similarity 0.88)
- NTClick: Achieving Precise Interactive Segmentation With Noise-tolerant Clicks (CVPR 2025 · similarity 0.86)
- Repurposing Stable Diffusion Attention for Training-Free Unsupervised Interactive Segmentation (CVPR 2025 · similarity 0.86)
- Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts (CVPR 2024 · similarity 0.85)
- TETRIS: Towards Exploring the Robustness of Interactive Segmentation (AAAI 2024 · similarity 0.85)
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.
