Research map / Detection and segmentation / CNN features
BDCN: Bi-Directional Cascade Network for Perceptual Edge Detection
Jianzhong He, Shiliang Steven Zhang, Ming Yang, Yanhu Shan, Tiejun Huang
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 shallow-to-deep cascade (ablation) | VGG16 baseline · BSDS500 val | ODS: .7681 -> .7683 | No clear effect |
| add deep-to-shallow cascade (ablation) | VGG16 baseline · BSDS500 val | AP: .7912 -> .8049 | Improved |
| add bi-directional cascade (S2D+D2S) (ablation) | VGG16 baseline · BSDS500 val | ODS: .7681 -> .7762 | Improved |
| add SEM (ablation) | VGG16 baseline · BSDS500 val | ODS: .7681 -> .7748 | Improved |
| add SEM on top of bi-directional cascade (ablation) | BDCN · BSDS500 val | ODS: .7762 -> .7765 | Improved |
| add SEM on top of bi-directional cascade (ablation) | BDCN · BSDS500 val | AP: .8013 -> .8091 | Improved |
| increase SEM dilated convolutions K from 0 to 3 (ablation) | SEM · BSDS500 val | ODS: .7728 -> .7748 | Improved |
| increase SEM convolutions K from 3 to 4 (ablation) | SEM · BSDS500 val | ODS: .7748 -> .7745 | No clear effect |
| use dilation rate factor r0=4 instead of 1,1,1 rates (ablation) | SEM · BSDS500 val | ODS: .7720 -> .7748 | Improved |
| increase dilation factor r0 from 4 to 8 (ablation) | SEM · BSDS500 val | ODS: .7748 -> .7742 | Hurt |
| use PASCAL-Context additional training data | BDCN · BSDS500 test | ODS: .806 -> .820 | Improved |
| use multi-scale fused inputs | BDCN · BSDS500 test | ODS: .820 -> .828 | Improved |
Later papers that compared with this method
6 later accepted papers report BDCN as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- Lightweight Pixel Difference Networks for Efficient Visual Representation Learning (TPAMI 2023 · better on 0 of 3 shared results)
- EDTER: Edge Detection With Transformer (CVPR 2022 · better on 5 of 5 shared results)
- Unmixing Convolutional Features for Crisp Edge Detection (TPAMI 2021 · better on 6 of 6 shared results)
- Pixel Difference Networks for Efficient Edge Detection (ICCV 2021 · better on 0 of 4 shared results)
- RINDNet: Edge Detection for Discontinuity in Reflectance, Illumination, Normal and Depth (ICCV 2021 · better on 1 of 1 shared result)
- Recurrent neural circuits for contour detection (ICLR 2020 · better on 0 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
- Bi-Directional Cascade Network for Perceptual Edge Detection (CVPR 2019 · similarity 0.99)
- ScaleNet - Improve CNNs through Recursively Rescaling Objects (AAAI 2020 · similarity 0.87)
- Feature Pyramid Networks for Object Detection (CVPR 2017 · similarity 0.87)
- Data-Driven Neuron Allocation for Scale Aggregation Networks (CVPR 2019 · similarity 0.87)
- A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection (ECCV 2016 · similarity 0.87)
- Harnessing Edge Information for Improved Robustness in Vision Transformers (AAAI 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.
