tutorials/video/wan/wan2-2-animate

[giscus[bot]]

tutorials/video/wan/wan2-2-animate

Unified character animation and replacement framework with precise motion and expression replication.

https://docs.comfy.org/tutorials/video/wan/wan2-2-animate

[marymistan]

Hi! What to do if this node (Video Sampling and output) is unknown when loading the workflow?

[comfyui-wiki]

What's your ComfyUI version? It needs ComfyUI 0.3.60. It's a subgraph.

[nonoDamn]

Hello,
‘’‘
For the “Video Extend” group, it’s in order to extend to the output video length
Each Video Extend will extend another 77 frames(Around 4.8125 seconds)
If your input video is less then 5s, you might not need it
If you want to extend longer, you need to copy and paste multiple times, you need to link the batch_images from last Video Extend to next one, and also the video_frame_offset from last Video Extend to next one
’‘’
I connected the nodes in this way, but the generated video is still only 4 seconds, and the multiple generated videos are not continuous. For example, if I input an 8s video, the first generated one is the first 2 seconds, the second generated one is the first 4 seconds, and the third generated one is the last 4 seconds. It's very strange. The ideal situation should be to generate a complete 8s video.

[comfyui-wiki]

Could you send the workflow here?

[nonoDamn]

Sorry for replying to you so late,
I solved this problem using the latest official workflow. It seems that there was an issue with the previous workflow

[PatrickBorkowicz]

For Move Mode, are you meant to disconnect the background_video and character_mask from all 'Video Extend' groups/subgraphs? I have an issue where each extended section of the video is progressively more zoomed in, like a jump cut.

Also not mentioned in the doc, but for extending I believe you are meant to re-active both the 'Video Extend' nodes and the subgraph (those nodes are bypassed in the template workflow.

[comfyui-wiki]

Yes, the extended part is bypassed by default; you need to enable it if you need it. If you want a longer duration, you have to copy and paste more extend groups.

[greek12man]

I tried to input a video with 1 frame duration, but the output was 1 minute 17 seconds, with that one image. Is there a way to limit the output to 1 frame?

[zerobell-lee]

This workflow has some flaws.

1. "Video extend" subgraph does not accept "height", so it is fixed to have same value as "width", resulting weird cropping.
2. "Video extend" accepts fps value separately, so if you don't sync them with main fps, it would lower fps suddenly.
3. Mix mode does not work properly. It remains black masks even after replaced original character. So it never looks natural.

[jjgarrid]

In cloud.confy.org using the default template I get the following error:

ComfyUI Error Report

Error Details

• Node ID: N/A
• Node Type: N/A
• Exception Type: ModelDownloadError
• Exception Message: unsupported model(s) for node 100 (type DWPreprocessor): [yolox_l.onnx]

Stack Trace

No stack trace available

System Information

• ComfyUI Version:
• Arguments:
• OS:
• Python Version:
• Embedded Python: false
• PyTorch Version:

Devices

Logs

[object Object]

Attached Workflow

Please make sure that workflow does not contain any sensitive information such as API keys or passwords.

Workflow too large. Please manually upload the workflow from local file system.

Additional Context

(Please add any additional context or steps to reproduce the error here)