Photo Background - 2d Compositing｜写真背景・二次元合成
Trained on 2d illustrations composited on a photo background.
This is a small LoRA I thought would be interesting to see how models trained on illustrations or real world images/video can produce the composite, mixed reality effect.
ℹ️ LoRA work best when applied to the base models on which they are trained. Please read the About This Version on the appropriate base models and workflow/training information.
Metadata is included in all uploaded files, you can drag the generated videos into ComfyUI to use the embedded workflows.
Description
FAQ
Comments (11)
Those example images you posted are absolutely beautiful!
I didn’t even know Anima had LoRA support yet, thanks for making this!
Yes some support now! It is still a preview model, so those lora probably wont work with the fully trained release.
is it work on preview2?
I LOVE IT， great lora.
anima
After looking at the example picture, it has little effect on the characters (which is good). If there could be a more extensive dataset (including various background elements), it would be preferable to have a version that does not specify trigger words and is solely based on natural language annotations.
Thanks, I noticed an issue with the new config and took down that version to retrain and test a little more.
The "photo background" tag isn't necessarily a trigger with how I've configured the captions, they do lean towards NL now and the base also learned this concept with that tag so should have a better effect if used.
@motimalu Circlestone Lab (Anima) have posted an official LoRa with their fuill training config for reference in case that helps: https://civitai.com/models/2536147/greg-rutkowski-style-anima
@deitychaser Thanks! Yes I was referencing that but made an error updating my workspace
@motimalu I see, cool, thank you. We will be waiting patiently.
Looking forward to the next release based on preview 0.3