Thanks for the step-by-step tutorial it helps a lot 🙏 After I did many tests via the online demo (huggingface) I found out that if you use MULTIPLE IMAGES of the same person with different lights, emotions, angles, the results will be MUCH more similar to the original. I'm very new to ComfyUI so I didn't figure out how to connect more than one node "Load Image" so the results are not as amazing as the online demo. I guess there is a way to load multiple source images of the same person to do the same on ComfyUI, please SHARE if it's a specific NODE. it will be nice to experiment. Thanks ahead! 🙏
Thank for watching! Yes you can add multiple images in ComfyUI by using the following steps. 1. Add a load image node and select the first image. 2. Connect this load image node with a batch image node. 3. Add a second image node and connect it to the batch image node. now you have two images from the batch image node. if you want to add more, then you connect the batch image node to another batch image node and keep chaining them. Hope this helps.
@yonathanevan these are the steps: search for Juggernaut XL on civitai website. The model id is 133005. Once the page loads, just below the name, there are different models you can choose from. Download the one that says: V9+RDPhoto2-Lightning_4S. I renamed mine to say ligntning_4_steps.
Hello, if the ""INSIGHTFACE_OPT" is not showing, it means you have an older version. Go into the ComfyUI Manager and click on Update All. Once the update completes, restart ComfyUI. Remove the old node and add a new one.
Hi if we do not use the output of image crop node, why do we use it at all? Also for some reason I am getting issues when i press Queue Prompt: Error occurred when executing PhotoMakerEncodePlus: integer division or modulo by zero File "E:\ComfyUI\execution.py", line 316, in execute output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb) File "E:\ComfyUI\execution.py", line 191, in get_output_data and it goes on bit more... Seems this workflow will fail to work with some images?
Hi, any tips how to solve this: "0.1 seconds (IMPORT FAILED): E:\software\ComfyUI\ComfyUI\custom_nodes\ComfyUI-PhotoMaker-Plus"? this is in CLI after ComfyUI restarted.
Hello, can you check your terminal and see if you have any "module not found" error. Usually, the import will fail if you are missing some dependencies. PhotoMaker will require installing insightface. If this is indeed the case, under the "Resources" section in the description of this video, there is a link "Watch for Insightface installation:". You can follow the instruction to get insightface and then try again.
@@CodeCraftersCorner managed to sort it out but only thanks to Claude and some tips from it. I wish installing and getting it to work was less hour consuming :) I had to among other things install parallel phyton version (same version as comfyui embedded python) so that I could copy include folder (where all .h files are), withouth this I could not compile insightface on windows, it complained that python could not find files in include folder). Is there comfyui installation that where embedded python comes with include folder? I wonder what kind of comfyui installation you have, since it appears you did not run into such an issue otherwise you would have mentioned it in the video :)
@@CodeCraftersCorner I figured it out, the tutorial will only work without issues and erros for properly installed ComfyUI with git clone, for portable, it won't work. User will run into issue with python not able to find all python sources etc, since embedded python that comes with protable is not complete python installation, so installing insightface and other stuff needed will lead to compile issues. Hope feedback helps
@UnchartedWorlds ah, yes. When installing ComfyUI with git clone, there are no python_embeded folder. Thank you for the feedback. For next videos, I will try to keep it in mind.
Thank you! Very detailed as always. Does anybody know what the standalone version uses for a model/checkpoint? I had good results with the standalone, but can't reproduce it in ComfyUI. What checkpoint is best? I tried EpicrealismXL but was not pleased. Also the Anime/Comic checkpoints don't work very well...
Thank you! I don't have any specific recommendations since I generate everything locally. I prefer not to suggest options I haven't personally tried and can't guarantee.
Thank you so much , you earned a sub with your explaination. TOp notch stuff
Thanks for the sub!
Simple, handy, to the point, Amazing guide for starting up the Photomaker v2
Great job!
Glad you enjoyed it!
Thanks for the step-by-step tutorial it helps a lot 🙏
After I did many tests via the online demo (huggingface) I found out that if you use MULTIPLE IMAGES of the same person with different lights, emotions, angles, the results will be MUCH more similar to the original.
I'm very new to ComfyUI so I didn't figure out how to connect more than one node "Load Image" so the results are not as amazing as the online demo.
I guess there is a way to load multiple source images of the same person to do the same on ComfyUI, please SHARE if it's a specific NODE. it will be nice to experiment.
Thanks ahead! 🙏
Thank for watching! Yes you can add multiple images in ComfyUI by using the following steps.
1. Add a load image node and select the first image.
2. Connect this load image node with a batch image node.
3. Add a second image node and connect it to the batch image node.
now you have two images from the batch image node.
if you want to add more, then you connect the batch image node to another batch image node and keep chaining them.
Hope this helps.
@@CodeCraftersCorner Thank you for your kind reply, I will give that a try later on 🙏
👍
@@CodeCraftersCorner Where I could fine the exact model of juggernaut? thanks
@yonathanevan these are the steps: search for Juggernaut XL on civitai website. The model id is 133005. Once the page loads, just below the name, there are different models you can choose from. Download the one that says: V9+RDPhoto2-Lightning_4S. I renamed mine to say ligntning_4_steps.
Thanks for the valulable information!
I've install everything like in the video,
but my "Photomaker Encode Plus" doesn't have input "INSIGHTFACE_OPT" .
Please help.... thanks
Hello, if the ""INSIGHTFACE_OPT" is not showing, it means you have an older version. Go into the ComfyUI Manager and click on Update All. Once the update completes, restart ComfyUI. Remove the old node and add a new one.
Hi if we do not use the output of image crop node, why do we use it at all? Also for some reason I am getting issues when i press Queue Prompt:
Error occurred when executing PhotoMakerEncodePlus:
integer division or modulo by zero
File "E:\ComfyUI\execution.py", line 316, in execute
output_data, output_ui, has_subgraph = get_output_data(obj, input_data_all, execution_block_cb=execution_block_cb, pre_execute_cb=pre_execute_cb)
File "E:\ComfyUI\execution.py", line 191, in get_output_data
and it goes on bit more...
Seems this workflow will fail to work with some images?
Yes, I also noticed that PhotoMaker V2 will throw errors when the input images are not "ideal".
can you help? Error occurred when executing PhotoMakerEncodePlus:
ZeroDivisionError: integer division or modulo by zero
Can you check if you have everything connected properly. The text should contain man img or woman img. The trigger word is img.
Hi, any tips how to solve this: "0.1 seconds (IMPORT FAILED): E:\software\ComfyUI\ComfyUI\custom_nodes\ComfyUI-PhotoMaker-Plus"? this is in CLI after ComfyUI restarted.
Hello, can you check your terminal and see if you have any "module not found" error. Usually, the import will fail if you are missing some dependencies. PhotoMaker will require installing insightface. If this is indeed the case, under the "Resources" section in the description of this video, there is a link "Watch for Insightface installation:". You can follow the instruction to get insightface and then try again.
@@CodeCraftersCorner managed to sort it out but only thanks to Claude and some tips from it. I wish installing and getting it to work was less hour consuming :) I had to among other things install parallel phyton version (same version as comfyui embedded python) so that I could copy include folder (where all .h files are), withouth this I could not compile insightface on windows, it complained that python could not find files in include folder). Is there comfyui installation that where embedded python comes with include folder? I wonder what kind of comfyui installation you have, since it appears you did not run into such an issue otherwise you would have mentioned it in the video :)
@@CodeCraftersCorner I figured it out, the tutorial will only work without issues and erros for properly installed ComfyUI with git clone, for portable, it won't work. User will run into issue with python not able to find all python sources etc, since embedded python that comes with protable is not complete python installation, so installing insightface and other stuff needed will lead to compile issues. Hope feedback helps
@UnchartedWorlds ah, yes. When installing ComfyUI with git clone, there are no python_embeded folder. Thank you for the feedback. For next videos, I will try to keep it in mind.
How will the sdxl PONY models work with this?
Not sure! If it is based off SDXL, then it may work.
Thank you! Very detailed as always.
Does anybody know what the standalone version uses for a model/checkpoint? I had good results with the standalone, but can't reproduce it in ComfyUI.
What checkpoint is best? I tried EpicrealismXL but was not pleased. Also the Anime/Comic checkpoints don't work very well...
Hello, The standalone project uses this model: 'SG161222/RealVisXL_V4.0'. You can search for it on Huggingface models.
Hi. I like your Videos. Do you have any Idea about the cheapest cloud GPU that is easy to use?
Thank you! I don't have any specific recommendations since I generate everything locally. I prefer not to suggest options I haven't personally tried and can't guarantee.
Try computegrid.ai it starts at 0.25$ an hour.