This repo is updated frequently, keeping up with the latest code is highly recommended.
2019.10.14The official PyTorch version of LFFD is first online. Now the repo is only preview version. Face detection code for v2 version is released nightly.2019.10.16Now the face detection code for v2 version can train normally. The code of other tasks will be updated soon.
This repo is the official PyTorch source code of paper "LFFD: A Light and Fast Face Detector for Edge Devices". Our paper presents a light and fast face detector (LFFD) for edge devices. LFFD considerably balances both accuracy and latency, resulting in small model size, fast inference speed while achieving excellent accuracy. Understanding the essence of receptive field makes detection networks interpretable.
In practical, we have deployed it in cloud and edge devices (like NVIDIA Jetson series and ARM-based embedding system). The comprehensive performance of LFFD is robust enough to support our applications.
In fact, our method is a general detection framework that applicable to one class detection, such as face detection, pedestrian detection, head detection, vehicle detection and so on. In general, an object class, whose average ratio of the longer side and the shorter side is less than 5, is appropriate to apply our framework for detection.
Several practical advantages:
- large scale coverage, and easy to extend to larger scales by adding more layers without much latency gain.
- detect small objects (as small as 10 pixels) in images with extremely large resolution (8K or even larger) in only one inference.
- easy backbone with very common operators makes it easy to deploy anywhere.
on the way
We re-implement the proposed method using PyTorch. The MXNet Version is here
- Python>=3.5
- numpy>=1.16 (lower versions should work as well, but not tested)
- PyTorch>=1.0.0 (install guide)
- cv2=3.x (pip3 install opencv-python==3.4.5.20, other version should work as well, but not tested)
Tips:
- use PyTorch with cudnn.
- build numpy from source with OpenBLAS. This will improve the training efficiency.
- make sure cv2 links to libjpeg-turbo, not libjpeg. This will improve the jpeg decode efficiency.
- face_detection contains the code of training, evaluation and inference for LFFD, the main content of this repo. The trained models of different versions are provided for off-the-shelf deployment.
- head_detection contains the trained models for head detection. The models are obtained by the proposed general one class detection framework.
- pedestrian_detection contains the trained models for pedestrian detection. The models are obtained by the proposed general one class detection framework.
- vehicle_detection contains the trained models for vehicle detection. The models are obtained by the proposed general one class detection framework.
- ChasingTrainFramework_GeneralOneClassDetection is a simple wrapper based on MXNet Module API for general one class detection.
- Download the repo:
git clone https://github.com/becauseofAI/lffd-pytorch.git
- Refer to the corresponding sub-project for detailed usage. Now only the v2 version of face_detection can be tried to train.
If you benefit from our work in your research and product, please kindly cite the paper
@inproceedings{LFFD,
title={LFFD: A Light and Fast Face Detector for Edge Devices},
author={He, Yonghao and Xu, Dezhong and Wu, Lifang and Jian, Meng and Xiang, Shiming and Pan, Chunhong},
booktitle={arXiv:1904.10633},
year={2019}
}
- face detection
- pedestrian detection
- head detection
- vehicle detection
- license plate detection
- reconstruction version
becauseofAI[1], Yonghao He[2]
[1]E-mails: helloai777@gmail.com
[2]E-mails: yonghao.he@ia.ac.cn / yonghao.he@aliyun.com
If you are interested in this work, any innovative contributions are welcome!!!
Internship is open at NLPR, CASIA all the time. Send me your resumes!