Research map / Detection and segmentation / CNN features
Deep Crisp Boundaries
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 the backward refinement pathway to HED (ablation) | CED · BSDS500 test | ODS: .780 -> .793 | Helped |
| Add sub-pixel convolution instead of deconvolution (ablation) | CED · BSDS500 test | ODS: .800 -> .803 | Helped |
| Add multi-scale testing (ablation) | CED · BSDS500 test | ODS: .794 -> .803 | Helped |
| Use the full refinement architecture | HED · BSDS500 test | ODS: .780 -> .803 | Helped |
| Augment training with VOC images | CED · BSDS500 test | ODS: .803 -> .815 | Helped |
| Use CED instead of HED edge maps for EpicFlow | EpicFlow · Sintel training set | AEE: 3.588 -> 3.570 | Helped |
| Use CED instead of HED as MCG's edge detector | MCG · VOC12 val set, top-100 proposals | mean Jaccard index (instance level): 64.9 -> 68.4 | Helped |
| Use CED instead of HED as MCG's edge detector | MCG · VOC12 val set, all proposals | mean Jaccard index (instance level): 80.8 -> 83.3 | Helped |
Closest papers by meaning
- DiffusionEdge: Diffusion Probabilistic Model for Crisp Edge Detection (AAAI 2024 · similarity 0.86)
- Harnessing Edge Information for Improved Robustness in Vision Transformers (AAAI 2024 · similarity 0.85)
- DECO: Unleashing the Potential of ConvNets for Query-based Detection and Segmentation (ICLR 2025 · similarity 0.84)
- Deep Edge Filter: Return of the Human-Crafted Layer in Deep Learning (NeurIPS 2025 · similarity 0.84)
- DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging (NeurIPS 2026 · similarity 0.84)
- MogaNet: Multi-order Gated Aggregation Network (ICLR 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.