Research map / Detection and segmentation / Weakly supervised detection and segmentation
Learning Saliency Propagation for Semi-Supervised Instance 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 |
|---|---|---|---|
| Add ShapeProp to Mask R-CNN | Mask R-CNN, ResNet50-FPN · COCO2017 non-VOC→VOC | AP: 23.9 -> 34.4 | Helped |
| Add ShapeProp to Mask R-CNN | Mask R-CNN, ResNet50-FPN · COCO2017 voc→non-voc | AP: 19.2 -> 30.4 | Helped |
| Add ShapeProp to Mask R-CNN | Mask R-CNN, ResNet101-FPN · COCO2017 non-VOC→VOC | AP: 24.7 -> 35.5 | Helped |
| Add ShapeProp to Mask R-CNN | Mask R-CNN, ResNet101-FPN · COCO2017 voc→non-voc | AP: 18.5 -> 31.9 | Helped |
| Train with additional box annotations | Mask R-CNN · BDD100K | AP: 21.6 -> 24.5 | Helped |
| Add ShapeProp | Mask R-CNN · BDD100K image-wise semi-supervision | AP: 24.5 -> 26.2 | Helped |
| Add ShapeProp | RetinaMask · BDD100K image-wise semi-supervision | AP: 24.4 -> 26.1 | Helped |
| Add ShapeProp | Mask R-CNN · BDD100K fully supervised | AP: 21.6 -> 22.7 | Helped |
| Add ShapeProp | Mask R-CNN · VOC12 fully supervised (1k images) | AP: 29.7 -> 30.6 | Helped |
| Add ShapeProp | Mask R-CNN · COCO17 fully supervised (120k images) | AP: 34.2 -> 35.7 | Helped |
| Increase mask-head channels to match ShapeProp parameter count (ablation) | Mask R-CNN · BDD100K | AP: 21.6 -> 21.4 | Hurt |
| Use channel-agnostic propagation weights (ablation) | ShapeProp with Mask R-CNN · BDD100K | AP: 22.7 -> 22.4 | Hurt |
Closest papers by meaning
- Instance-Level Salient Object Segmentation (CVPR 2017 · similarity 0.89)
- Classifying, Segmenting, and Tracking Object Instances in Video with Mask Propagation (CVPR 2020 · similarity 0.88)
- Learning to Segment Every Thing (CVPR 2018 · similarity 0.88)
- S4Net: Single Stage Salient-Instance Segmentation (CVPR 2019 · similarity 0.87)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · similarity 0.86)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (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.