PyramidBox++: High Performance Detector for Finding Tiny Face

Abstract

With the rapid development of deep convolutional neural network, face detection has made great progress in recent years. WIDER FACE dataset, as a main benchmark, contributes greatly to this area. A large amount of methods have been put forward where PyramidBox designs an effective data augmentation strategy (Data-anchor-sampling) and context-based module for face detector. In this report, we improve each part to further boost the performance, including Balanced-data-anchor-sampling, Dual-PyramidAnchors and Dense Context Module. Specifically, Balanced-data-anchor-sampling obtains more uniform sampling of faces with different sizes. Dual-PyramidAnchors facilitate feature learning by introducing progressive anchor loss. Dense Context Module with dense connection not only enlarges receptive filed, but also passes information efficiently. Integrating these techniques, PyramidBox++ is constructed and achieves state-of-the-art performance in hard set.

References in corpus (7)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Focal Loss for Dense Object Detection
• DenseBox: Unifying Landmark Localization with End to End Object Detection
• IntroVAE: Introspective Variational Autoencoders for Photographic Image Synthesis
• Acquisition of Localization Confidence for Accurate Object Detection
• SFace: An Efficient Network for Face Detection in Large Scale Variations
• Seeing Small Faces from Robust Anchor's Perspective

Cited by in corpus (12)

• TinaFace: Strong but Simple Baseline for Face Detection
• Going Deeper Into Face Detection: A Survey
• YOLO5Face: Why Reinventing a Face Detector
• Learning Independent Instance Maps for Crowd Localization
• Accurate Face Detection for High Performance
• Crowd Localization from Gaussian Mixture Scoped Knowledge and Scoped Teacher
• A survey of face recognition techniques under occlusion
• RefineFace: Refinement Neural Network for High Performance Face Detection
• Face Detection with Feature Pyramids and Landmarks
• LDC-Net: A Unified Framework for Localization, Detection and Counting in Dense Crowds
• Disentangle Your Dense Object Detector
• MOS: A Low Latency and Lightweight Framework for Face Detection, Landmark Localization, and Head Pose Estimation