Research map / Detection and segmentation / Open-set and few-shot detection
WIDER FACE: A Face Detection Benchmark
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 |
|---|---|---|---|
| Train ACF on WIDER FACE rather than use the external-data baseline | ACF · WIDER FACE | average AP: 5.4% | Helped |
| Train Faceness on WIDER FACE rather than use the external-data baseline | Faceness · WIDER FACE Hard | AP: 4.2% | Helped |
| Train ACF on WIDER FACE rather than use the external-data baseline | ACF · FDDB | recall: 87.48%; 1.4% | Helped |
| Train Faceness on WIDER FACE rather than use the external-data baseline | Faceness · FDDB | recall: 91.78%; 0.8% | Helped |
| Use multiple networks specialized for different scale ranges instead of one network for multiple scales | Two-stage CNN · WIDER FACE Hard | AP: 8.5% | Helped |
| Use multiple networks specialized for different scale ranges instead of one network for multiple scales | Two-stage CNN · WIDER FACE Easy | AP | No clear effect |
| Train on WIDER FACE rather than the external-data baseline | ACF and Faceness · WIDER FACE event categories | AP: 56 out of 60 | Helped |
Closest papers by meaning
- MogFace: Towards a Deeper Appreciation on Face Detection (CVPR 2022 · similarity 0.88)
- Finding Tiny Faces (CVPR 2017 · similarity 0.88)
- The MegaFace Benchmark: 1 Million Faces for Recognition at Scale (CVPR 2016 · similarity 0.88)
- CRFace: Confidence Ranker for Model-Agnostic Face Detection Refinement (CVPR 2021 · similarity 0.86)
- GigaHumanDet: Exploring Full-Body Detection on Gigapixel-Level Images (AAAI 2024 · similarity 0.84)
- Vec2Face: Scaling Face Dataset Generation with Loosely Constrained Vectors (ICLR 2025 · similarity 0.84)
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.