I'm experiencing several issues with ComfyUI_RopeWrapper when processing videos:
Partial video processing: The face swapping only works for the first few seconds of a video, not the entire duration, even when frames are properly loaded.
Inconsistent behavior: When adding new videos to the workflow, the face swapping sometimes stops working entirely.
"Stack expects a non-empty TensorList" error: This error appears during processing, causing the operation to fail.
Environment
OS: Ubuntu 22.04.5 LTS
GPU: NVIDIA RTX A6000
RAM: 96GB
CPU: Intel Xeon E5-2683 v4 @ 2.10GHz
ComfyUI: Latest version
Required extensions: VideoHelperSuite and KJNodes installed
Using inswapper_128.onnx model as recommended for Linux
I'm experiencing several issues with ComfyUI_RopeWrapper when processing videos:
Partial video processing: The face swapping only works for the first few seconds of a video, not the entire duration, even when frames are properly loaded.
Inconsistent behavior: When adding new videos to the workflow, the face swapping sometimes stops working entirely.
"Stack expects a non-empty TensorList" error: This error appears during processing, causing the operation to fail.
Environment
OS: Ubuntu 22.04.5 LTS
GPU: NVIDIA RTX A6000
RAM: 96GB
CPU: Intel Xeon E5-2683 v4 @ 2.10GHz
ComfyUI: Latest version
Required extensions: VideoHelperSuite and KJNodes installed
Using inswapper_128.onnx model as recommended for Linux