Research map / Detection and segmentation / CNN features
Semantic Segmentation With Boundary Neural Fields
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 |
|---|---|---|---|
| predict boundaries from interpolated semantic FCN feature maps | BNF boundary detector · BSDS500 | AP: 0.833 -> 0.851 | Helped |
| replace RGB affinity with semantic boundary-based affinity | global inference methods on SBD · SBD | PP-IOU: 74.6 -> 77.6 | Helped |
| replace RGB affinity with semantic boundary-based affinity | global inference methods on SBD · SBD | PI-IOU: 53.0 -> 58.5 | Helped |
| use boundary-based affinity instead of RGB affinity | Belief Propagation on SBD · SBD | PP-IOU: 75.4 -> 75.6 | Helped |
| use boundary-based affinity instead of RGB affinity | ICM on SBD · SBD | PI-IOU: 45.7 -> 48.8 | Helped |
| use boundary-based affinity instead of RGB affinity | TRWS on SBD · SBD | PP-IOU: 75.9 -> 76.7 | Helped |
| use boundary-based affinity instead of RGB affinity | QPBO on SBD · SBD | PI-IOU: 55.3 -> 57.2 | Helped |
| use BNF global inference with boundary and softmax affinities instead of Dense-CRF | semantic segmentation on SBD · SBD | PP-IOU: 77.3 -> 77.6 | Helped |
| use BNF global inference with boundary and softmax affinities instead of QPBO | semantic segmentation on SBD · SBD | PI-IOU: 57.2 -> 58.5 | Helped |
| add softmax-based affinities to boundary-based affinities (ablation) | BNF-SB on SBD · SBD | PP-IOU: 77.4 -> 77.6 | Helped |
| add softmax-based affinities to boundary-based affinities (ablation) | BNF-SB on SBD · SBD | PI-IOU: 58.5 -> 58.5 | No clear effect |
| use the proposed global inference instead of QPBO with BNF affinity | semantic segmentation on SBD · SBD | PP-IOU: 77.2 -> 77.6 | Helped |
Closest papers by meaning
- Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation (CVPR 2016 · similarity 0.91)
- Semantic Diffusion Network for Semantic Segmentation (NeurIPS 2022 · similarity 0.91)
- Semantic Image Segmentation With Task-Specific Edge Detection Using CNNs and a Discriminatively Trained Domain Transform (CVPR 2016 · similarity 0.90)
- Learning Fully Dense Neural Networks for Image Semantic Segmentation (AAAI 2019 · similarity 0.89)
- Convolutional Random Walk Networks for Semantic Image Segmentation (CVPR 2017 · similarity 0.89)
- Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling (CVPR 2024 · similarity 0.88)
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.