Sign In

Generative AI Style Transfer - QuantArt

Updated: Aug 12, 2024

backgroundphotorealistic

Download

1 variant available

fp32 PickleTensor

last.ckpt

Full precision, largest file • 509.65 MB

Verified:

You need these files to run this model. We'll show the best match for your preferences.

Config

test.yaml1.83 KB

Verified:

Downloads your preferred variants

Type
Checkpoint Trained
Stats

51

Reviews
Published

Aug 12, 2024

Base Model

SD 1.5

Hash
AutoV2
07F52C0147
default creator card background decoration
Followers - 4

4

Likes - 8

8

2.jpg

The project is about generative AI style transfer. Mainly, it aims to build an interface and UI for landscape to art style transfer using a method called QuantArt (QuantArt: Quantizing Image Style Transfer Towards High Visual Fidelity). The QuantArt model produces impressive results, particularly for landscape to art transformations, and exhibits strong generalization capabilities. The project's objective is to provide easy accessibility, allowing it to be utilized as part of other tech solutions like generating YouTube thumbnails, visuals for stories, and covers for posts. The project includes a layer built on top of the original GitHub repo, incorporating different models that utilize the QuantArt mechanism.