Home

Awesome

DBFace | 中文说明 | ncnn支持

DBFace is a real-time, single-stage detector for face detection, with faster speed and higher accuracy

MAP@0.5 Result on validation set of WiderFace

MethodVersionSizeEasyMediumHard
RetinaFace-MobileNetV2Small1.68MB0.8960.8710.681
DBFace-Small-H-NoExt(Ours)Small1.30MB0.8950.8700.713
DBFace-Small-H(Ours)Small1.73MB0.8990.8760.728
DBFace-MobileNetV3 (Ours)Large7.03MB0.9050.8960.794
CenterFace-MobileNetV2Large7.3MB???

Result on DBFace (threshold = 0.2)

selfie


Result on RetinaFace-MobileNetV2 (threshold=0.2)

retinaface

Result on CenterFace-MobileNetV2 (threshold=?)

selfie.centerface.draw.jpg

MAP@0.5 Result on validation set of WiderFace

Train

train.md

<br/>

Onnx And TensorRT

tensorRTIntegrate

<br/>

QQ Group

936842116

<br/>

Author

References

  1. Hamid Rezatofighi1, Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regression:https://arxiv.org/abs/1902.09630

  2. Xingyi Zhou, Objects as Points:https://arxiv.org/abs/1904.07850

  3. Zili Liu, Training-Time-Friendly Network for Real-Time Object Detection:https://arxiv.org/abs/1909.00700

  4. Zhen-Hua Feng, Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks: https://arxiv.org/abs/1711.06753v4

  5. Mahyar Najib, SSH: Single Stage Headless Face Detector: https://arxiv.org/abs/1708.03979

  6. MobileNet: https://github.com/xiaolai-sqlai/mobilenetv3