-
This repo's origin code from (https://github.com/TreB1eN/InsightFace_Pytorch)
-
python 3.6
-
nvidia driver : 410.78
-
cuda : 10.0
-
pytorch : 1.4.0
Pre-trained model is here Google drive
Mobilefacenet and Ir_se
set the model configuration in config.py line 19
conf.use_mobilfacenet = True
if True, you will use mobliefacenet or else
Provide the face images your want to detect in the data/face_bank folder, and guarantee it have a structure like following:
data/facebank/
---> id1/
---> id1_1.jpg
---> id2/
---> id2_1.jpg
---> id3/
---> id3_1.jpg
---> id3_2.jpg
I prepare facebank for news and parasite videos. please name it as facebank
The videos are not provided
model file name as 'final_model' for GPU or 'final_model_cpu' for CPU
provided pre trained models are GPU version
```
python infer_on_video.py -f [video file name] -s [save file name]
```
the video file should be inside the data/face_bank folder
- Video Detection Demo @Youtube
change face detection network