Talking-Face-Generation-DAVS icon indicating copy to clipboard operation
Talking-Face-Generation-DAVS copied to clipboard

result file

Open TahoeWang opened this issue 5 years ago • 2 comments

I got two kinds of photos as result: test_sample1_fake_audio_B_0_x.png and test_sample1_real_A_x.png. what do they mean? And how to use them to get the final video?

TahoeWang avatar Jan 17 '20 09:01 TahoeWang

Using test_sample1_fake_audio_B_0_x.png to get the final video, test_sample1_real_A_x.png is the real image, test_sample1_fake_audio_B_0_x.png is synthesized by gan.

ZhengMengbin avatar Mar 26 '20 11:03 ZhengMengbin

@ZhengMengbin hi, I have a question! How to get the synthesized video using test_sample1_fake_B_0_x.png? The number of image sequence is 99, according to the process of preprocessing.py, one generated image should corresponds to 5 video sequence, hence, when generate the final video image should copy 5. Thus the generated video with 25 fps has 20 seconds long, while the duration of the test audio sample is 7 seconds. How to properly generate the final video? Thanks a lot !

lovekittynine avatar Sep 27 '21 09:09 lovekittynine