Research map / Detection and segmentation / CNN features
Optimal Proposal Learning for Deployable End-to-End Pedestrian Detection
Xiaolin Song, Binghui Chen, Pengyu Li, Jun-Yan He, Biao Wang, Yifeng Geng, Xuansong Xie, Honggang Zhang
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 |
|---|---|---|---|
| replace FCOS with OPL | pedestrian detection · CrowdHuman val | mMR: 45.6 -> 44.9 | Improved |
| replace FCOS with OPL | pedestrian detection · CrowdHuman val | AP: 90.7 -> 91.0 | Improved |
| replace FCOS with OPL | pedestrian detection · TJU-Ped-campus | mMR (RS): 62.7 -> 61.7 | Improved |
| replace FCOS with OPL | pedestrian detection · TJU-Ped-traffic | mMR (A): 39.7 -> 38.7 | Improved |
| replace FCOS with OPL | pedestrian detection · Caltech test | mMR (R): 6.9 -> 5.2 | Improved |
| add C2F and CPN (ablation) | FCOS one-to-one detector · CrowdHuman val | mMR: 49.3 -> 44.9 | Improved |
| add C2F (ablation) | FCOS one-to-one detector · CrowdHuman val | mMR: 49.3 -> 47.2 | Improved |
| add CPN (ablation) | FCOS one-to-one detector · CrowdHuman val | mMR: 49.3 -> 47.0 | Improved |
| use two C2F refinement steps with M={16,4} (ablation) | OPL with CPN · CrowdHuman val | mMR: 47.0 -> 44.9 | Improved |
| use C2F-1step with M={4} (ablation) | OPL with CPN · CrowdHuman val | mMR: 47.0 -> 46.1 | Improved |
| use C2F-1step with M={9} (ablation) | OPL with CPN · CrowdHuman val | mMR: 47.0 -> 45.8 | Improved |
| use C2F-1step with M={16} (ablation) | OPL with CPN · CrowdHuman val | mMR: 47.0 -> 46.4 | 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.
- FCOS — FCOS: Fully Convolutional One-Stage Object Detection (ICCV 2019) · better on 3 of 3 shared results
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
- What Can Help Pedestrian Detection? (CVPR 2017 · similarity 0.86)
- Where, What, Whether: Multi-Modal Learning Meets Pedestrian Detection (CVPR 2020 · similarity 0.86)
- Bi-box Regression for Pedestrian Detection and Occlusion Estimation (ECCV 2018 · similarity 0.86)
- Mask-Guided Attention Network for Occluded Pedestrian Detection (ICCV 2019 · similarity 0.85)
- Semi-Supervised Pedestrian Instance Synthesis and Detection With Mutual Reinforcement (ICCV 2019 · similarity 0.85)
- Is Faster R-CNN Doing Well for Pedestrian Detection? (ECCV 2016 · 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.
