Open woct0rdho opened 3 months ago
Am I correct in that the right most image in the row of 3 is your IC-light output and then the final is essentially a composite of high frequency + IC light?
The top left one is the original image, the top center one is the low frequency of the original image (so the difference between the two is the high frequency), and the top right one is the low frequency of the processed image
The code does not have the subtraction and addition part, can you add it please?
The code does not have the subtraction and addition part, can you add it please?
We have another (rather trivial) script here, and in real use there can be more pipeline between the two scripts
@woct0rdho Nice results! Did you use a foreground cutout pipeline from here or a pipeline from here, or you write another pipeline?
Hello! Can you show how you apply the code with LC Lights? It is not clear at what step you use the results obtained by cutting out the background and applying its embeddings in t2i/i2i pipelines
Nice work! I'm already using it in our indie game dev. I see some people complaining that the AI will change details in the original image, so I just want to mention that we can separate the high-frequency component from the original image and replace it into the processed image.
We can use, e.g. Erasing Appearance Preservation to separate the base color and the high-frequency component, for which I have a Python implementation here