magic-research / magic-avatar

MagicAvatar: Multimodal Avatar Generation and Animation
https://magic-avatar.github.io/
BSD 3-Clause "New" or "Revised" License
618 stars 33 forks source link

Code Release #1

Open EatEmAll opened 1 year ago

EatEmAll commented 1 year ago

Are there any plans to release the code & pretrained checkpoints?

jfzhang95 commented 1 year ago

Hi, thanks so much for your interest in our work! Our code is currently under review as per company policy. We will release it as soon as possible.

diviner2 commented 1 year ago

看了预告,很想体验一下,希望尽快发布

liuhongen1234567 commented 1 year ago

想看一下文本如何生成动作的,其他部分感觉text2video-zero 之类的模型就可以实现。

avenger123456 commented 1 year ago

同想看文本如何生成这么连贯的motion的

jfzhang95 commented 1 year ago

Thanks for your interest in our work. You can refer to MDM for text-to-motion generation.

basicbh commented 1 year ago

哈哈,字节就是不给你,你没办法,期待他们不是骗子。

makhnanov commented 11 months ago

Most Wanted Code =)

blx0102 commented 10 months ago

Under review forever

Echolink50 commented 10 months ago

They released magic animate which is similar and if magic animate is any indication the actual results won't be as good as the examples for the initial release.