sjtuplayer/few-shot-diffusion

[ICCV 2023] Phasic Content Fusing Diffusion Model with Directional Distribution Consistency for Few-Shot Model Adaption

22
/ 100
Experimental

This tool helps artists, designers, and content creators adapt an existing image generation model to produce new images in a distinct style, even with very few examples of that style. You provide a small set of images representing a new style (like "Van Gogh faces" or "cartoons"), and the tool processes these along with a base model (like one trained on human faces). The output is a refined model capable of generating new images that blend the content of the base model with the aesthetic of your few examples.

No commits in the last 6 months.

Use this if you need to generate images in a highly specific style or domain but only have a handful of example images to learn from.

Not ideal if you have a large dataset of images in your target style, as traditional training methods might be more straightforward.

generative art style transfer digital content creation image synthesis design prototyping
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 8 / 25
Maturity 8 / 25
Community 6 / 25

How are scores calculated?

Stars

67

Forks

3

Language

Python

License

Last pushed

Dec 07, 2023

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/diffusion/sjtuplayer/few-shot-diffusion"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.