Tensorflow-object-detection-API-for-face-detcetion

use tensorflow object detection API to detect face. This project base on tf object detection API and use wider face dataset. You can build a object detection project with Tf offical example(object_detection_tutorial.ipynb) step by step. You can clone this project and training your own dataset or detect custom picture without any other setting.

1. clone project

git clone https://github.com/aodiwei/Tensorflow-object-detection-API-for-face-detcetion.git

2. install necessary python libary

• tensorflow
• Cython
• opencv

3. bulid dataset

• download wider dataset from http://mmlab.ie.cuhk.edu.hk/projects/WIDERFace/
• data_process/dataset_util.py can help you transfer wider format data to tf record, cause tf API only support tf-record
• data_process/boundingbox.py can help you understand wider format coordinate to tf record coordinate

4. download pre-train model

download from (http://download.tensorflow.org/models/object_detection/ssd_mobilenet_v1_coco_2017_11_17/.tar.gz) and extract to pre_train_ckp/ssd_mobilenet_v1_coco_11_06_2017

5. training

with default args:

python3 train.py 

with custom args

python3 train.py \
        --logtostderr \
        --train_dir=path/to/train_dir \
        --pipeline_config_path=pipeline_config.pbtxt

6.eval

python3 eval.py

7. export

export model with export_inference_graph.py

8. detection

You can use my trained model to detect custom image simply

python3 detection_cv.py --input=your_img.jpg

more details please read detection_cv.py

tf offical detection example in detection.py

result sample:

9. Citation

@inproceedings{yang2016wider,
Author = {Yang, Shuo and Luo, Ping and Loy, Chen Change and Tang, Xiaoou},
Booktitle = {IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
Title = {WIDER FACE: A Face Detection Benchmark},
Year = {2016}}

https://github.com/priya-dwivedi/Deep-Learning/tree/master/tensorflow_toy_detector
https://github.com/yeephycho/widerface-to-tfrecord