Talking Face Video Generation using Deep Learning
I had the opportunity to do a research internship at INRIA, a french research institute. I was working on talking face video generation using GANs and CNNs.
Video puppeteering, on the left the driving video and on the right the generated video
I wrote an article where I summarized state-of-the-art research in video generation. I then compared them on the same data. When the code wasn’t available I reimplemented the code or part of the code in PyTorch.
You can find the article here.
I then implemented a new archtitecture based on CNNs and a GAN developed at INRIA. Unfortunately, I didn’t have the time to test it.
You can find the code on github here.
I also collected a new audio-video dataset using selenium that I automatically cropped around the face.