Research map / Detection and segmentation / CNN features
Part-Aware Context Network for Human Parsing
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 |
|---|---|---|---|
| use PCNet instead of DPC | human parsing · PASCAL-Person-Part | mIoU: 71.34 -> 74.59 | Helped |
| use PCNet instead of BraidNet | human parsing · LIP | mIoU: 54.4 -> 57.03 | Helped |
| use PCNet instead of BraidNet | human parsing · CIHP | mIoU: 60.62 -> 61.05 | Helped |
| add a part class module (ablation) | ResNet-101 human parsing · PASCAL-Person-Part val | mIoU: 66.61 -> 67.40 | Helped |
| use channel-wise multiplication for relational aggregation (ablation) | PCNet with part class module · PASCAL-Person-Part val | mIoU: 67.40 -> 68.53 | Helped |
| replace channel-wise multiplication with dynamic convolution in relational aggregation (ablation) | PCNet · PASCAL-Person-Part val | mIoU: 68.53 -> 69.41 | Helped |
| use three rather than two graph convolution layers (ablation) | PCNet relational aggregation module · PASCAL-Person-Part val | mIoU: 70.12 -> 70.52 | Helped |
| use dynamic convolution rather than concatenation in relational dispersion (ablation) | PCNet · PASCAL-Person-Part val | mIoU: 71.08 -> 72.26 | Helped |
| add multi-scale input data augmentation during training (ablation) | PCNet · PASCAL-Person-Part val | mIoU: 72.26 -> 73.32 | Helped |
| add multi-scale testing (ablation) | PCNet · PASCAL-Person-Part val | mIoU: 73.32 -> 74.59 | Helped |
Closest papers by meaning
- Hierarchical Human Parsing With Typed Part-Relation Reasoning (CVPR 2020 · similarity 0.91)
- CDGNet: Class Distribution Guided Network for Human Parsing (CVPR 2022 · similarity 0.90)
- Progressive Cognitive Human Parsing (AAAI 2018 · similarity 0.90)
- Progressive One-shot Human Parsing (AAAI 2021 · similarity 0.89)
- Look Into Person: Self-Supervised Structure-Sensitive Learning and a New Benchmark for Human Parsing (CVPR 2017 · similarity 0.89)
- Semantic Human Parsing via Scalable Semantic Transfer Over Multiple Label Domains (CVPR 2023 · similarity 0.89)
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.