John Gettings comments

Results 4 comments of


                                            John Gettings

Higher resolution pretrained model ?

Check out the --upscale argument. Default is 2 so your image is initially 256x256. Change that to 4 and output will be 1024x1024. Results are great too. It utilizes RealESRGAN

Higher resolution pretrained model ?

Just try --upscale 4. It's what you're looking for. I think they use realESRGAN for the background and something else for the face. I don't remember. It comes out 1024x1024...

How can I use gfpgan for video?

Video to frames: ``` def vid2frames(vidPath, framesOutPath): vidcap = cv2.VideoCapture(vidPath) success,image = vidcap.read() frame = 1 while success: cv2.imwrite(framesOutPath + str(frame).zfill(5) + '.png', image) success,image = vidcap.read() frame += 1...

Any examples on real human picture rather than gan generated one?

Check out the deepfake example video. https://m.youtube.com/watch?v=nPAV-jpTzqI Though I think the stylgan2 faces look best. I don't think real people will look more realistic. Check out the examples colab. It...