search

HumanAIGC / EMO

Emote Portrait Alive: Generating Expressive Portrait Videos with Audio2Video Diffusion Model under Weak Conditions
7.4k stars 898 forks source link

An alternative project EchoMimic is open-sourced #271

Open lymhust opened 2 months ago

lymhust commented 2 months 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 image

https://github.com/HumanAIGC/EMO/assets/18476076/8f091654-713e-4e41-9ad1-0a2a0322d01f

https://github.com/HumanAIGC/EMO/assets/18476076/f65e8542-b101-45bd-8a08-734badc7fc18

JoeFannie commented 2 months ago

It is a powerful tool to realize "talking head" generation