Research map / Detection and segmentation / CNN features
FA-RPN: Floating Region Proposals for Face 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 |
|---|---|---|---|
| Reduce anchor stride from 32 to 16 pixels at inference (ablation) | FA-RPN trained with stride 32 · WIDER | detection performance: 3.8% | Helped |
| Add an inference-time anchor refinement step (ablation) | FA-RPN with denser anchors · WIDER | detection performance: 1.4% | Helped |
| Use scale-normalized multi-scale training with SNIP (ablation) | FA-RPN-based Faster R-CNN detector · WIDER | detection performance: 1% | Helped |
| Initialize ResNet-50 with COCO detection pre-training (ablation) | FA-RPN-based face detector · WIDER | detection performance | Helped |
| Use iterative refinement by pooling again on regressed proposals (ablation) | FA-RPN · WIDER | proposal precision and recall | Helped |
Closest papers by meaning
- Region Proposal by Guided Anchoring (CVPR 2019 · similarity 0.87)
- Cascade RPN: Delving into High-Quality Region Proposal Network with Adaptive Convolution (NeurIPS 2019 · similarity 0.86)
- R-FCN++: Towards Accurate Region-Based Fully Convolutional Networks for Object Detection (AAAI 2018 · similarity 0.86)
- HyperNet: Towards Accurate Region Proposal Generation and Joint Object Detection (CVPR 2016 · similarity 0.86)
- R-FCN: Object Detection via Region-based Fully Convolutional Networks (NeurIPS 2016 · similarity 0.85)
- DeRPN: Taking a Further Step toward More General Object Detection (AAAI 2019 · 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.