face_blur

Detects and blurs any faces in a given image or video.

Dependencies

• Python3
• OpenCV
• Tensorflow
• MTCNN

Install Dependencies

pip3 install -r ./requirements.txt

Instructions

face_blur utilizes predefined arguments upon execution. Input (-i) and output (-o) are both required. Resize image (-r) and skip frame (-s) are both optional. All video and image formats are supported.

Blur an Image

Use the input argument (-i) to specify the image to blur. Then use the output argument (-o) to specify a write to file.

python3 face_blur.py -i image.jpg -o image_blurred.jpg

Resize an Image

The input image can be resized to improve performance during processing. Use the resize argument (flagged as -r) to shrink the image. The argument value is the percent change in size.

python3 face_blur.py -i image.jpg -o image_resized_blurred.jpg -r 25

Blur a Video

Use the input argument (-i) to specify the video to blur. Then use the output argument (-o) to specify a write to file.

python3 face_blur.py -i video.mp4 -o video_blurred.mp4

Skip Frames

Frames can be skipped to improve performance during processing. Use the skip frame argument (flagged as -s) to reduce the number of frames for processing. The argument value indicates how frequently to skip a frame.

python3 face_blur.py -i video.mp4 -o video_skipped_blurred.mp4 -s 10