Privateness is all the time extremely essential, particularly with visible media the place it’s possible you’ll not have the permission of people within the video. In the event you’re filming one thing in public, it is possible you will catch somebody’s face who merely would not need or must be recognized. This just lately bought me to pondering: what is the best strategy to blur faces in a video by way of command line?

The perfect open supply utility I discovered for blurring faces in a video was deface. Let’s take a look at how you should utilize deface to blur faces in movies!

Begin by downloading Python-based by way of pip:

python3 -m pip set up deface

With deface put in, merely present the video title and get the output file with blurred faces:

sudo deface ./sample-4k-faces-video.mp4
Enter:  ./sample-4k-faces-video.mp4
Output: ./sample-4k-faces-video_anonymized.mp4
100%|█████████████████████████████

The ensuing video does a formidable job of blurring out faces of individuals strolling by within the unique recording:

View the ensuing video of individuals strolling down the streets of New York:

The default threshold for face recognition works very nicely, even on transferring topics. You’ll be able to experiment with thresholds with the thresh argument, and even draw the thresholds out whereas debugging:

I downloaded a handful of YouTube movies utilizing my favourite YouTube downloading utility youtube-dl and I used to be amazed at how nicely deface did on quite a lot of visible environments. Faces have been recognized at a dependable stage even at default threshold!