Research map / Detection and segmentation / CNN features
What Can Help Pedestrian Detection?
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 |
|---|---|---|---|
| Jointly learn segmentation prediction and detection | HyperLearner · KITTI validation, 1x input | Moderate: 69.80 -> 71.15 | Helped |
| Jointly learn edge prediction and detection | HyperLearner · KITTI validation, 2x input | Moderate: 71.73 -> 72.51 | Helped |
| Add person segmentation supervision | Faster R-CNN · Cityscapes validation, 540p input | AP: 74.97 -> 77.22 | Helped |
| Add person segmentation supervision | Faster R-CNN · Cityscapes validation, 720p input | AP: 86.89 -> 87.67 | Helped |
| Add segmentation channel as an input | Faster R-CNN · KITTI validation, 1x input | Moderate: 68.96 -> 70.80 | Helped |
| Add heatmap channel as an input | Faster R-CNN · KITTI validation, 1x input | Moderate: 68.96 -> 70.33 | Helped |
| Add ICF channel as an input | Faster R-CNN · KITTI validation, 1x input | Moderate: 68.96 -> 68.40 | Hurt |
| Add optical-flow channel as an input | Faster R-CNN · KITTI validation, 1x input | Moderate: 68.96 -> 69.39 | Helped |
| Add heatmap channel as an input | Faster R-CNN · KITTI validation, 2x input | Moderate: 71.21 -> 71.39 | No clear effect |
| Use two side-branch convolution layers instead of one (ablation) | Faster R-CNN with extra channel features · KITTI validation | Moderate: 70.40 -> 70.80 | Helped |
| Initialize the side branch with pretrained weights instead of random weights (ablation) | Faster R-CNN with extra channel features · KITTI validation | Moderate: 70.80 -> 69.92 | Hurt |
| Remove CFN channel-feature supervision (ablation) | HyperLearner · KITTI validation, 1x input | Moderate: 71.15 -> 69.72 | Hurt |
Closest papers by meaning
- Where, What, Whether: Multi-Modal Learning Meets Pedestrian Detection (CVPR 2020 · similarity 0.89)
- High-Level Semantic Feature Detection: A New Perspective for Pedestrian Detection (CVPR 2019 · similarity 0.88)
- Generalizable Pedestrian Detection: The Elephant in the Room (CVPR 2021 · similarity 0.88)
- Occluded Pedestrian Detection Through Guided Attention in CNNs (CVPR 2018 · similarity 0.87)
- Pedestrian Detection With Autoregressive Network Phases (CVPR 2019 · similarity 0.87)
- How Far Are We From Solving Pedestrian Detection? (CVPR 2016 · similarity 0.86)
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.