Research map / Detection and segmentation / CNN features
Adaptive Pyramid Context Network for 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 multi-scale adaptive context with scales {1,2,3,6} to dilated ResNet50 FCN (ablation) | Dilated ResNet50 FCN · PASCAL VOC 2012 validation | mIoU: 69.83 -> 78.20 | Helped |
| Use pyramid scales {1,2,3,6,32} instead of {1,2,3,6} (ablation) | APCNet with ResNet50 · PASCAL VOC 2012 validation | mIoU: 78.20 -> 77.29 | Hurt |
| Add global-guided local affinity (GLA) (ablation) | APCNet with ResNet50 · PASCAL VOC 2012 validation | mIoU: 77.68 -> 78.20 | Helped |
| Add global-guided local affinity (GLA) (ablation) | APCNet with ResNet101 · PASCAL VOC 2012 validation | mIoU: 80.17 -> 80.71 | Helped |
| Replace ResNet50 with ResNet101 (ablation) | APCNet with scales {1,2,3,6} · PASCAL VOC 2012 validation | mIoU: 78.20 -> 80.71 | Helped |
| Add deep supervision (ablation) | APCNet with ResNet101 · PASCAL VOC 2012 validation | mIoU: 80.71 -> 80.93 | Helped |
| Add horizontal-flip evaluation (ablation) | APCNet with deep supervision · PASCAL VOC 2012 validation | mIoU: 80.93 -> 81.33 | Helped |
| Add multiscale evaluation (ablation) | APCNet with deep supervision and flip evaluation · PASCAL VOC 2012 validation | mIoU: 81.33 -> 81.93 | Helped |
| Fine-tune on the original PASCAL VOC 2012 training set (ablation) | APCNet with deep supervision, flip, and multiscale evaluation · PASCAL VOC 2012 validation | mIoU: 81.93 -> 82.67 | Helped |
| Add an adaptive context module (ablation) | PSPNet reimplementation · PASCAL VOC 2012 validation | mIoU: 79.79 -> 80.19 | Helped |
| Add global-guided local affinity to the adaptive context module (ablation) | PSPNet reimplementation with adaptive context · PASCAL VOC 2012 validation | mIoU: 80.19 -> 80.93 | Helped |
| Use APCNet instead of EncNet | Semantic segmentation models · PASCAL VOC 2012 test | mIoU: 82.9 -> 84.2 | Helped |
Closest papers by meaning
- Context Encoding for Semantic Segmentation (CVPR 2018 · similarity 0.89)
- Context-Reinforced Semantic Segmentation (CVPR 2019 · similarity 0.89)
- Context Prior for Scene Segmentation (CVPR 2020 · similarity 0.88)
- Pyramid Scene Parsing Network (CVPR 2017 · similarity 0.87)
- Contextrast: Contextual Contrastive Learning for Semantic Segmentation (CVPR 2024 · similarity 0.86)
- Test-Time Adaptation of Vision-Language Models for Open-Vocabulary Semantic Segmentation (NeurIPS 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.