Closed lymhust closed 1 month ago
EchoMimic is capable of generating portrait videos not only by audios and facial landmarks individually, but also by a combination of both audios and selected facial landmarks. Project link: https://badtobest.github.io/echomimic.html GitHub link: https://github.com/BadToBest/EchoMimic
https://github.com/X-LANCE/AniTalker/assets/18476076/d3f6b8f8-a8ee-4332-b586-85bcfd3cc701
https://github.com/X-LANCE/AniTalker/assets/18476076/83b7dd63-0e24-4b1b-84de-8655ddb9a3b7
Wow~ I've also run this code. It's excellent, worth following.
but its vey slow, more then 2 hours of GPU time for 30 secs video generation, not useful for practical application.
EchoMimic is capable of generating portrait videos not only by audios and facial landmarks individually, but also by a combination of both audios and selected facial landmarks. Project link: https://badtobest.github.io/echomimic.html GitHub link: https://github.com/BadToBest/EchoMimic
https://github.com/X-LANCE/AniTalker/assets/18476076/d3f6b8f8-a8ee-4332-b586-85bcfd3cc701
https://github.com/X-LANCE/AniTalker/assets/18476076/83b7dd63-0e24-4b1b-84de-8655ddb9a3b7