Research map / Architectures and efficiency / Attention and transformers
Probabilistic Attention for Interactive Segmentation
Prasad Gabbur, Manjot Bilkhu, Javier Movellan
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 |
|---|---|---|---|
| Adapt keys online from observed queries | Probabilistic BoTNet50 · GrabCut and Berkeley, low-feedback regime | mIoU: ~10% absolute improvement without corrective clicks | Improved |
| Use key adaptation with lower prior precision, including maximum-likelihood adaptation (ablation) | Probabilistic BoTNet50-FactoredPE · GrabCut and Berkeley | mIoU | Improved |
| Use full rather than factored relative position encoding (ablation) | Probabilistic BoTNet50 · GrabCut and Berkeley | mIoU | Improved |
| Replace standard backbone attention with probabilistic attention | BoTNet50 · GrabCut and Berkeley, especially with fewer clicks | mIoU | Improved |
| Propagate fixed annotator-provided values online | Corrective self-attention classification head · GrabCut and Berkeley, high-feedback regime | mIoU | Improved |
| Use key adaptation and value propagation together | Probabilistic BoTNet50 with corrective self-attention head · GrabCut and Berkeley, low- and high-feedback regimes | mIoU | Improved |
| Use value propagation in both axial-attention height and width blocks instead of only the width block (ablation) | Corrective self-attention classification head · Interactive segmentation | mIoU | Hurt |
| Use value propagation for multiple iterations (1 or 5 iterations were tested) (ablation) | Corrective self-attention classification head · GrabCut and Berkeley | mIoU | Improved |
Closest papers by meaning
- Attention (as Discrete-Time Markov) Chains (NeurIPS 2025 · similarity 0.86)
- Repurposing Stable Diffusion Attention for Training-Free Unsupervised Interactive Segmentation (CVPR 2025 · similarity 0.85)
- Rethinking Interactive Image Segmentation with Low Latency High Quality and Diverse Prompts (CVPR 2024 · similarity 0.85)
- Order-aware Interactive Segmentation (ICLR 2025 · similarity 0.84)
- Dissecting Query-Key Interaction in Vision Transformers (NeurIPS 2024 · similarity 0.84)
- Density Matters: Improved Core-Set for Active Domain Adaptive Segmentation (AAAI 2024 · similarity 0.84)
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.
