Error When Loading 4000x4000 Image Size in LoadImage Node

[fengxu11]

When attempting to load an image of 4000x4000 dimensions, I receive an error message.

I remember using this node in November 2023 without any errors. How should I solve this problem?

• Loading: ComfyUI-Manager (V2.9)
• ComfyUI Revision: 2046 [a38b9b3a] | Released on '2024-03-04'

Here is the error message:

Error occurred when executing LoadImage:

Sizes of tensors must match except in dimension 0. Expected size 4000 but got size 1080 for tensor number 1 in the list.


File "/root/ComfyUI/execution.py", line 152, in recursive_execute
output_data, output_ui = get_output_data(obj, input_data_all)
File "/root/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 "/root/ComfyUI/execution.py", line 75, in map_node_over_list
results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
File "/root/ComfyUI/nodes.py", line 1502, in load_image
output_image = torch.cat(output_images, dim=0)

Here is my workflow:

• This is the simplest workflow

{
  "last_node_id": 12,
  "last_link_id": 11,
  "nodes": [
    {
      "id": 10,
      "type": "LoadImage",
      "pos": [
        260,
        340
      ],
      "size": [
        320,
        310
      ],
      "flags": {},
      "order": 0,
      "mode": 0,
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            10
          ],
          "shape": 3,
          "label": "图像"
        },
        {
          "name": "MASK",
          "type": "MASK",
          "links": null,
          "shape": 3,
          "label": "遮罩"
        }
      ],
      "properties": {
        "Node name for S&R": "LoadImage"
      },
      "widgets_values": [
        "11_原 (1).JPG",
        "image"
      ]
    },
    {
      "id": 11,
      "type": "ImageInvert",
      "pos": [
        680,
        510
      ],
      "size": {
        "0": 210,
        "1": 26
      },
      "flags": {},
      "order": 1,
      "mode": 0,
      "inputs": [
        {
          "name": "image",
          "type": "IMAGE",
          "link": 10,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "outputs": [
        {
          "name": "IMAGE",
          "type": "IMAGE",
          "links": [
            11
          ],
          "shape": 3,
          "label": "图像"
        }
      ],
      "properties": {
        "Node name for S&R": "ImageInvert"
      }
    },
    {
      "id": 12,
      "type": "PreviewImage",
      "pos": [
        990,
        530
      ],
      "size": {
        "0": 210,
        "1": 26
      },
      "flags": {},
      "order": 2,
      "mode": 0,
      "inputs": [
        {
          "name": "images",
          "type": "IMAGE",
          "link": 11,
          "label": "图像",
          "slot_index": 0
        }
      ],
      "properties": {
        "Node name for S&R": "PreviewImage"
      }
    }
  ],
  "links": [
    [
      10,
      10,
      0,
      11,
      0,
      "IMAGE"
    ],
    [
      11,
      11,
      0,
      12,
      0,
      "IMAGE"
    ]
  ],
  "groups": [],
  "config": {},
  "extra": {},
  "version": 0.4
}

[YYH2913]

Same issue. I conducted a separate test for this module, and it appears that the issue is not with comfyui but with the Pillow library. In my tests, when the size of a JPEG exceeds 4096*4096, Pillow incorrectly processes the image twice and scales it down. I will submit this issue along with my test code there.

[YYH2913]

Same issue. I conducted a separate test for this module, and it appears that the issue is not with comfyui but with the Pillow library. In my tests, when the size of a JPEG exceeds 4096*4096, Pillow incorrectly processes the image twice and scales it down. I will submit this issue along with my test code there.

https://github.com/python-pillow/Pillow/issues/7954

[Fashizel]

I downgraded to 10.2.0 and that fixed the issue for now