BeautifAI - Image Upscaler & Enhancer - How-to

Поделиться
HTML-код
  • Опубликовано: 21 дек 2024

Комментарии • 33

  • @RealmsByLP
    @RealmsByLP 8 месяцев назад +1

    Great workflow. This video really helped me set it up! I've never gotten a custom workflow to work 😅

  • @archibible
    @archibible 8 месяцев назад

    This is awesome mate!! Big congrats on the new workflow 💪🏼🙌🏼🔥

  • @royandescartes
    @royandescartes 9 месяцев назад

    that workflow is super huge wow

  • @sorijin
    @sorijin 8 месяцев назад

    How did I miss this!? So dope I gotta try this out!

  • @huwhitememes
    @huwhitememes 6 месяцев назад

    Huge contribution to the community. Even my "so so" generations look great with this! 😂😎👌🏻

    • @midjourneyman
      @midjourneyman  6 месяцев назад +1

      So glad to read that!!! :)

    • @huwhitememes
      @huwhitememes 6 месяцев назад

      @@midjourneyman Kudos, Brother 🙏🏻😎

    • @midjourneyman
      @midjourneyman  6 месяцев назад +1

      @@huwhitememes Brad and I love it too :)

  • @SyntaxDiffusion
    @SyntaxDiffusion 9 месяцев назад

    EPIC!

  • @ShimruAttur
    @ShimruAttur 5 месяцев назад +2

    I am getting the following error:
    Error occurred when executing ImageResize+:
    'bool' object has no attribute 'startswith'
    File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\ComfyUI_windows_portable\ComfyUI\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    File "C:\ComfyUI_windows_portable\ComfyUI\custom_nodes\ComfyUI_essentials\image.py", line 295, in execute
    elif method.startswith('fill'):
    ^^^^^^^^^^^^^^^^^
    What could be the reason?

    • @renderartist
      @renderartist 4 месяца назад +2

      Got the same exact error, won't run without that resolved in the Resolution Calculator node group.

  • @Ai_mayyit
    @Ai_mayyit 7 месяцев назад

    Very amazing today i will try this method and please make a video on animate diff

  • @ericcarlson8864
    @ericcarlson8864 4 месяца назад

    I have the manger but everytime I go to install the custom nodes it never gets everything >_

  • @Vigilence
    @Vigilence 8 месяцев назад +2

    If I want to upscale the 4k image to 8k, do I change the upscale settings? It seems to stay at 4k.

    • @midjourneyman
      @midjourneyman  8 месяцев назад +1

      I'll test but the factors for pass 1 and 2 should help you achieve that.

  • @Imhotchep
    @Imhotchep 6 месяцев назад

    Hello,
    Thanks for the great work. Need help. Im getting this error : Error occurred when executing ImageResize+:
    'bool' object has no attribute 'startswith'
    The image resize gets a red corner. But when I change the method to anything but "true", the red corners are gone but I get this error then : Error occurred when executing CLIPVisionLoader:'NoneType' object has no attribute 'lower'
    I check anywhere I don't see any red corners
    Any help will be greatly appreciated

    • @midjourneyman
      @midjourneyman  6 месяцев назад

      Mmmmh, that is a strange one, did you try to google the error to see if it brings something useful up?

    • @life_of_luis
      @life_of_luis 6 месяцев назад

      I used a workflow from someone else and the AnimateDiff Loader node had some name similar to the models I used. You have to make sure that every models that are selected are the same path as yours by selecting the node and choosing the path in the node

  • @keepitrollin3670
    @keepitrollin3670 8 месяцев назад

    Looks kickass, but i'm getting a "insightface model is required for FaceID models" error message. Any ideas on how to fix it?
    I noticed that the workflow changed a bit since you made this video.

    • @SuperBeastsAI
      @SuperBeastsAI 8 месяцев назад +2

      I've seen this and I'm not sure which module is calling it but it doesn't stop the workflow from running.

    • @keepitrollin3670
      @keepitrollin3670 8 месяцев назад

      @@SuperBeastsAI thanks for a quick reply. For me it was stopping the whole process. After swapping the pairing beteween ipadapter model and lora, it's running fine. thanks!

    • @marcdevinci893
      @marcdevinci893 4 месяца назад

      @@keepitrollin3670 I get the same error, can't quite get your explanation to fix it. Could you please elaborate, it's a big workflow! cheers

  • @mcselcik
    @mcselcik 8 месяцев назад

    Great work but i keep getting "Can't import color-matcher ERROR". Eventhough installed all the requirements. Any idea why?

    • @midjourneyman
      @midjourneyman  8 месяцев назад

      That's odd! I am not sure why, but you could try to run it without the color match nodes to see if it goes through.

  • @TheGreatJ
    @TheGreatJ 9 месяцев назад

    Hey making the tutorial is super helpful! Thank you. I just wanted to ask. I am having bit of confusion on where to put the control net models. I downloaded the tile model and all 4 of the other control net models you linked (it seems like you are only using canny and depth). Is there a specific place that I need to place those models? I currently have a .bin model inside the control net folder that is being used for face_adapterv2... But I thought that the only files that were supposed to go in the control net folder were .bin types and that .safetensors were supposed to go somewhere else?
    I would appreciate the clarification if you could... Thanks for the amazing work!

    • @SuperBeastsAI
      @SuperBeastsAI 9 месяцев назад

      Hey @TheGreatJ yeah the 3 Control net models in the workflow are the Tile, Depth and Canny models
      .Safetensors can definitely be placed in the controlnet folder :)
      Because the filenames of some of these files is sometimes pretty bad I would suggest creating subfolders e.g. 'SDXL' in the model specific directory. Comfy UI has access to any subfolders as well and will show these when you refresh your ComfyUI.
      So you can then can put your SDXL Control net models specifically in /models/controlnet/SDXL
      So in our case this should include ttplanetSDXLControlnet_v10F16.safetensors, t2i-adapter_diffusers_xl_depth_zoe.safetensors and t2i-adapter_diffusers_xl_canny.safetensors
      Similarly I would suggest the same for the primary model E.g juggernaut goes into models/checkpoints/SDXL and the Upscale model goes into models/upscale_models (These aren't specific to SD versions so it can just be at the base directory).
      Keep in mind the way everyone stores their files is different so for every new workflow you need to load it up and then reselect each of these models based on the location/filename you actually used. Good luck!

    • @midjourneyman
      @midjourneyman  9 месяцев назад

      I see Brad answered. Please let me know if you have more questions.

  • @P4TCH5S
    @P4TCH5S 9 месяцев назад +2

    dude u made a tutorial!!!!