Under Construction
2262
Architecture (Cut Ver.)DJ Myosuke & Noizenecio
mapped by Faputa
submitted
last updated
2:51
300
1,764
2
HP Drain
6
Accuracy
8
Star Rating
8.90

Description

Collab with Cynply.
Image generated from VQGAN and CLIP.

VQGAN is a generative adversarial neural network that is good at generating images that look similar to others.

CLIP is another neural network that is able to determine how well a caption (or prompt) matches an image.

Check out this paper for more information https://arxiv.org/pdf/2204.08583.pdf

Success Rate

14.6%

Points of Failure

/

stick figure