kijai / ComfyUI-MimicMotionWrapper

Apache License 2.0
313 stars 25 forks source link

The number of poses must be greater than the context size #68

Open sofknk00l opened 2 months ago

sofknk00l commented 2 months ago

using comfy UI portable getting this error

Error occurred when executing MimicMotionSampler:

The number of poses must be greater than the context size

File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 152, in recursive_execute output_data, output_ui = get_output_data(obj, input_data_all) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 82, in get_output_data return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI_windows_portable\ComfyUI\execution.py", line 75, in map_node_over_list results.append(getattr(obj, func)(**slice_dict(input_data_all, i))) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-MimicMotionWrapper\nodes.py", line 287, in process assert B >= context_size, "The number of poses must be greater than the context size" ^^^^^^^^^^^^^^^^^

joanygaard commented 2 months ago

Getting same error.

kijai commented 2 months ago

What exactly is unclear about the error?

joanygaard commented 2 months ago

I'm new to this, I don't know what context_size is, what kind of unit is it? Pixels? Frames? In my case I think the problem is lack of working workflow examples showing how the image input is wired. Seems like some re-arrangement of custom node modules has caused most workflow examples to become invalid. My animation is 56 PNGs. Reference PNG and pose PNGs are 512x768 pixels. If the processing expects an image list list but instead get image one-by-one, the error is understandable I guess if context_size is a unit of frames.

joanygaard commented 2 months ago

I'm new to this, I don't know what context_size is, what kind of unit is it? Pixels? Frames? In my case I think the problem is lack of working workflow examples showing how the image input is wired. Seems like some re-arrangement of custom node modules has caused most workflow examples to become invalid. My animation is 56 PNGs. Reference PNG and pose PNGs are 512x768 pixels. If the processing expects an image list list but instead get image one-by-one, the error is understandable I guess if context_size is a unit of frames.

I managed to make it work by using a video/gif input source instead of single batched images.

MihaiRotariu commented 4 weeks ago

If I am not mistaken, the context_size value of the MimicMotion Sampler node has to be lower than the number of frames in your video(or the number of frames from your video you are trying to process). Try lowering that.

motoyagugu commented 4 weeks ago

If I am not mistaken, the context_size value of the MimicMotion Sampler node has to be lower than the number of frames in your video(or the number of frames from your video you are trying to process). Try lowering that^^^^^ My Frame Cap is 60, but the context is 72, so I encounter with this error, adjust it to 60, it works.