Research map / Detection and segmentation / CNN features
Multi-Scale Bidirectional FCN for Object Skeleton Extraction
Read the paperCheck 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 multi-scale bidirectional architecture to ResNet-FCN (ablation) | ResNet-FCN · Sym-PASCAL | F-measure: 0.515 -> 0.557 | Helped |
| add dense connections to MSB (ablation) | ResNet-FCN+MSB · Sym-PASCAL | F-measure: 0.557 -> 0.570 | Helped |
| use ResNet-101 dilated FCN baseline instead of VGG-based SRN | skeleton detection · Sym-PASCAL | F-measure: 0.443 -> 0.515 | Helped |
| single-stage end-to-end network instead of detector followed by FSDS | skeleton detection · Sym-PASCAL | F-measure: 0.354 -> 0.570 | Helped |
| use only deep features (pyramid pooling) instead of skip-layer shallow features | skeleton detection · SK-SMALL | F-measure: 0.632 -> 0.738 | Helped |
| MSB-FCN vs SRN inference speed | MSB-FCN · Sym-PASCAL | runtime (s): 0.12 -> 0.13 | No clear effect |
Closest papers by meaning
- Object Skeleton Extraction in Natural Images by Fusing Scale-Associated Deep Side Outputs (CVPR 2016 · similarity 0.90)
- Bi-Directional Cascade Network for Perceptual Edge Detection (CVPR 2019 · similarity 0.86)
- DeepFlux for Skeletons in the Wild (CVPR 2019 · similarity 0.86)
- Feature Pyramid Networks for Object Detection (CVPR 2017 · similarity 0.86)
- Fully Convolutional Network Based Skeletonization for Handwritten Chinese Characters (AAAI 2018 · similarity 0.84)
- Attention to Scale: Scale-Aware Semantic Image Segmentation (CVPR 2016 · 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.