You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hello I've no interest in installing this locally just want to play with this tool for simple alterations of images so I've followed the huggingface demo but it seems I get an error every time. https://huggingface.co/spaces/Shitao/OmniGen
I upload 2 images and simply ask the ai to replace the clothing item of image 1 with image 2 using this prompt:
replace the woman's shoes in <|image_1|> with the shoes of <|image_2|>
*The source images are very small only 800x500 on average.
*I've even tried setting image output to as low as 256 but still error.
The text was updated successfully, but these errors were encountered:
Well first the image tags should be <img><|image_n|></img>, not <|image_n|>.
The demo currently runs on custom hardware as far as I'm aware, and is said to be quite unstable at some times due to quota limit. If you got a machine capable enough, you might want to run it locally instead of relying on the online demo. I believe that with some research it would even be possible to run on colab's free tier. For that you might want to check out this pull request #151.
Hello I've no interest in installing this locally just want to play with this tool for simple alterations of images so I've followed the huggingface demo but it seems I get an error every time.
https://huggingface.co/spaces/Shitao/OmniGen
I upload 2 images and simply ask the ai to replace the clothing item of image 1 with image 2 using this prompt:
replace the woman's shoes in <|image_1|> with the shoes of <|image_2|>
*The source images are very small only 800x500 on average.
*I've even tried setting image output to as low as 256 but still error.
The text was updated successfully, but these errors were encountered: