comfyui-liveportrait

comfyui-liveportrait copied to clipboard

I need a way to avoid crashes due to lack of CPU RAM. Could you create an argument to help with this? For example, Reactor and LivePortrait don't use my...

Amit30swgoh

OSError: [Errno 22] Invalid argument

1

writing 1008 Dump to K:\ComfyUI-aki-v1\temp\live_portrait_00000.mp4 #scenes_video [0] ['K:\\ComfyUI-aki-v1\\temp\\live_portrait_00000.mp4'] total_frame_count 1008 Using fallback file for extremely long metadata: 7919798/32567 !!! Exception during processing !!! [Errno 22] Invalid argument Traceback (most recent...

robji15

High-quality zero-shot lipsync pipeline built on LivePortrait

2

Hey folks! My team has been exploring zero-shot lipsyncing for a bit and we think we've improved on MuseTalk's quality quite a bit by using LivePortrait to neutralize expression and...

mvoodarla

V2V的计算机制似乎不正确，流程用时实在太长了

1

尝试运行了297帧（原视频和驱动视频帧率一致）。等了30分钟都没有结果。GPU占用3G，再等下去感觉遥遥无期。。。

Eikwang

Error occurred when executing ExpressionEditor_: invalid load key, '\x10'.

Abhijeetisilanos

ImportError: cannot import name 'compute_eye_delta' from 'LivePortrait.src.utils.retargeting_utils'

1

 I am new here, and I have an unexpected error when starting ComfyUI. Thanks so much for helping me to solve this problem.

irrikrlla

我个人的一点想法，因为我是做三维绑定的，所以看到这个插件后，我就想能不能把表情分的更细。 比如眼睛分左右，眉毛分左右，嘴角分左右等，其他能够分左右的都分开控制。 之后可能这个节点的属性太多了， 需不需要把ExpressionEditor 节点拆分为 眼睛 嘴巴 眉毛 3部分。 还有鼻子，眼珠，瞳孔等，其他口型啥的也可以考虑下。

yuzehong1986

多人表情变化不明显

1

请问多人场景下源视频人脸变化夸张，但转到多人后仅眼部有明显变化，嘴和脸部变化微弱 需要调整哪些参数 https://github.com/user-attachments/assets/1a8cd036-05e6-4c2e-9cef-2d8551805728

g711ab

@shadowcz007 你好，Live Portrait节点是否可以支持，特定项表情系数置零， 比如我不希望驱动视频头部晃动传递给图像

arixlin

gaoxuxu110