Bugfix for using nodes in video creation, where one source image is used.

[FortunaCournot]

Bugfix for:

File "...\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI-ReActor\nodes.py", line 1470, in execute result[image_index] = pasting * paste_mask + result[image_index] * (1. - paste_mask) ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~ RuntimeError: The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2

[Buumcode]

where one source image is used

Can you explain - what do you mean "one source image used"? If we do face swap im2video - we always has one Source image and one Target video. And with using VHS Video Helper Suite workflow - I never get any errors.

[FortunaCournot]

At ReActor FastFace Swap: InputImage from VAE Decode. This is almost the same.
But i have another node ReActor 🌌 Masking Helper, like in your other workflow, directly behind.

[FortunaCournot]

I removed the Masking Helper node. Now the video looks correct. My workflow was wrong, i havn't seen your example. Thank you!!! However, it is still a bug ;-)
Side note: I don't use this for vid2vid. I need it for img2vid, to keep character consistent over multiple segments.

[sahibalejandro]

This issue happens to me too, it only happens when the input_image is a video and the ReActor Masking Helper node is before the Video Combine (the last node).

[Buumcode]

Yes, I'm got this error too - with using Masking Helper:

got prompt
[ReActor] 23:44:21 - STATUS - Working: source face index [0], target face index [0]
[ReActor] 23:44:21 - STATUS - Analyzing Source Image...
[ReActor] 23:44:21 - STATUS - Analyzing Target Image...
[ReActor] 23:44:22 - STATUS - Swapping...
[ReActor] 23:44:25 - STATUS - --Done!--

0: 384x640 1 face, 87.2ms
Speed: 2.8ms preprocess, 87.2ms inference, 11.5ms postprocess per image at shape (1, 3, 384, 640)
Moving to GPU... OK
!!! Exception during processing !!! The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2
Traceback (most recent call last):
  File "c:\ComfyUI\execution.py", line 498, in execute
    output_data, output_ui, has_subgraph, has_pending_tasks = await get_output_data(prompt_id, unique_id, obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                                                              ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "c:\ComfyUI\execution.py", line 316, in get_output_data
    return_values = await _async_map_node_over_list(prompt_id, unique_id, obj, input_data_all, obj.FUNCTION, allow_interrupt=True, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb, hidden_inputs=hidden_inputs)
                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "c:\ComfyUI\execution.py", line 290, in _async_map_node_over_list
    await process_inputs(input_dict, i)
  File "c:\ComfyUI\execution.py", line 278, in process_inputs
    result = f(**inputs)
             ^^^^^^^^^^^
  File "C:\ComfyUI\custom_nodes\comfyui-reactor-node\nodes.py", line 1358, in execute
    result[image_index] = pasting * paste_mask + result[image_index] * (1. - paste_mask)
                                                 ~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~
RuntimeError: The size of tensor a (3) must match the size of tensor b (4) at non-singleton dimension 2
```

[Gourieff]

This is because Masking Helper requires list of images as input if you process a batch: