Research map / Detection and segmentation / CNN features
Object-Contextual Representations for Semantic Segmentation
Yuhui Yuan, Xilin Chen, Jingdong Wang
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 supervision on the soft object regions (ablation) | OCR (ResNet-101) · Cityscapes val | mIoU: 77.31% -> 79.58% | Improved |
| compute pixel-region relations from pixel and region representations instead of pixel only (DA scheme) (ablation) | OCR · Cityscapes val | mIoU: 79.01% -> 79.58% | Improved |
| compute relations from pixel and region representations instead of an intermediate segmentation map (ACF scheme) (ablation) | OCR · Cityscapes val | mIoU: 78.02% -> 79.58% | Improved |
| use OCR instead of PPM | dilated ResNet-101 · Cityscapes test (w/o coarse) | mIoU: 80.3% -> 81.8% | Improved |
| use OCR instead of ASPP | dilated ResNet-101 · Cityscapes test (w/o coarse) | mIoU: 81.0% -> 81.8% | Improved |
| use OCR instead of PPM | dilated ResNet-101 · ADE20K val | mIoU: 44.50% -> 45.28% | Improved |
| use OCR instead of Self Attention | dilated ResNet-101 · LIP val | mIoU: 55.15% -> 55.60% | Improved |
| use OCR instead of Double Attention | dilated ResNet-101 · Cityscapes test (w/o coarse) | mIoU: 81.2% -> 81.8% | Improved |
| use HRNetV2-W48 backbone instead of ResNet-101 with OCR | OCR · Cityscapes test (w/o coarse) | mIoU: 81.8% -> 82.4% | Improved |
| apply PPM or ASPP on HRNet-W48 (ablation) | HRNet-W48 · Cityscapes | mIoU | Hurt |
| add OCR to Panoptic-FPN | Panoptic-FPN (ResNet-101) · COCO val 2017 panoptic | PQ: 43.0% -> 44.2% | Improved |
| add OCR to Panoptic-FPN | Panoptic-FPN (ResNet-50) · COCO val 2017 panoptic | PQ: 41.5 -> 42.7 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DANet — DANet: Divergent Activation for Weakly Supervised Object Localization (ICCV 2019) · better on 1 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
- OCNet: Object Context for Semantic Segmentation (IJCV 2021 · similarity 0.93)
- Mining Contextual Information Beyond Image for Semantic Segmentation (ICCV 2021 · similarity 0.92)
- ISNet: Integrate Image-Level and Semantic-Level Context for Semantic Segmentation (ICCV 2021 · similarity 0.90)
- Self-Supervised Learning of Object Parts for Semantic Segmentation (CVPR 2022 · similarity 0.89)
- Distilling Spectral Graph for Object-Context Aware Open-Vocabulary Semantic Segmentation (CVPR 2025 · similarity 0.88)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.87)
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.
