Research map / Detection and segmentation / CNN features
Co-Occurrent Features in Semantic 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 ACF Module to FCN (ablation) | FCN on PASCAL-Context · PASCAL-Context | mIoU: 46.3 -> 49.8 | Helped |
| Add global pooling feature to ACF (ablation) | CFNet on PASCAL-Context · PASCAL-Context | mIoU: 49.8 -> 51.6 | Helped |
| Add Context Encoding Module (ablation) | CFNet on PASCAL-Context · PASCAL-Context | mIoU: 51.6 -> 52.4 | Helped |
| Use ResNet-101 instead of ResNet-50 (ablation) | CFNet on PASCAL-Context · PASCAL-Context | mIoU: 52.4 -> 54.9 | Helped |
| Use a contextual prior with K=2 rather than K=1 at H=1 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 49.4 -> 51.6 | Helped |
| Increase contextual-prior mixtures from K=2 to K=4 at H=1 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 51.6 -> 51.8 | Helped |
| Increase heads from H=1 to H=2 at K=1 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 49.4 -> 49.4 | No clear effect |
| Increase heads from H=1 to H=2 at K=2 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 51.6 -> 51.8 | Helped |
| Increase heads from H=1 to H=2 at K=4 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 51.8 -> 51.6 | Hurt |
| Increase heads from H=1 to H=4 at K=1 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 49.4 -> 51.2 | Helped |
| Increase heads from H=1 to H=4 at K=2 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU: 51.6 -> 52.1 | Helped |
| Increase heads from H=1 to H=4 at K=4 (ablation) | ACF on PASCAL-Context · PASCAL-Context | mIoU | Mixed |
Closest papers by meaning
- Separate and Conquer: Decoupling Co-occurrence via Decomposition and Representation for Weakly Supervised Semantic Segmentation (CVPR 2024 · similarity 0.87)
- Context Encoding for Semantic Segmentation (CVPR 2018 · similarity 0.87)
- Deep Object Co-Segmentation via Spatial-Semantic Network Modulation (AAAI 2020 · similarity 0.86)
- Unsupervised Semantic Segmentation Through Depth-Guided Feature Correlation and Sampling (CVPR 2024 · similarity 0.85)
- COS3D: Collaborative Open-Vocabulary 3D Segmentation (NeurIPS 2025 · similarity 0.85)
- Integrating Low-Level Visual Cues for Enhanced Unsupervised Semantic Segmentation (AAAI 2025 · similarity 0.85)
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.