v1.1 Update: Looks like I didn't test it enough because v1.0 looks pretty bad. Re-trained v1.1 after pruning some of the dataset and with more steps. This version should be less blobby and closer to the Mirror's Edge aesthetic.
XL LoRA trained on Mirror's Edge 1 screenshots. It's an update of an old dreambooth model but should be more flexible.
Trigger words: Mirror's Edge style, cityscape, abstract world, sterile, clean
Description
Looks like I didn't test it enough because v1.0 looks pretty bad. Re-trained v1.1 after pruning some of the dataset and with more steps. This version should be less blobby and closer to the Mirror's Edge aesthetic.
FAQ
Comments (4)
Could you redo this for SD1.5?
I plan to later when I work on the dataset again.
Love to see someone appreciate the Mirror's Edge aesthetic as much as I do. Thank you for making this!
Thanks a ton for uploading this. The solarpunk/modernist aesthetic of Mirror's Edge lends its self perfectly to image generation whether it'd be for SD or DALL-E. Wish there was more of this.