mtcnn-opencv

No Caffe required on your machine for MTCNN inferencing with this implementation. OpenCV uses DNN module that provides the inference support.

The module is capable of taking models & weights from various popular frameworks such as Caffe, tensorflow, darknet etc.

More info here - https://github.com/opencv/opencv/wiki/Deep-Learning-in-OpenCV

MTCNN

[ZHANG2016] Zhang, K., Zhang, Z., Li, Z., and Qiao, Y. (2016). Joint face detection and alignment using multitask cascaded convolutional networks. IEEE Signal Processing Letters, 23(10):1499–1503. https://kpzhang93.github.io/MTCNN_face_detection_alignment/paper/spl.pdf

Quickstart

Requirements

• OpenCV 4.1+
• CMake 3.2+

Build

git clone https://github.com/egcode/mtcnn-opencv.git
cd mtcnn-opencv
mkdir build
cd build
cmake ..
make

Run

cd mtcnn-opencv

# Camera Inference example (<app_binary> <path_to_models_dir>):
./build/infer_cam ./models


# Photo Inference examples (<app_binary> <path_to_models_dir> <path_to_test_image>):

# An image with 10 human faces
./build/infer_photo ./models ./data/got.jpg 

# An image with 46 human faces
./build/infer_photo ./models ./data/marvel.jpg

Results

Here is an example of how the execution of the sample application looks like

Acknowledgments

The model files are taken from here

10 human faces picture taken from here here

46 human faces picture taken from here here

References

https://github.com/ksachdeva/opencv-mtcnn

https://github.com/golunovas/mtcnn-cpp