Skip to content

Latest commit

 

History

51 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 

Repository files navigation

CALDMTIS (Comparative Analysis of Latent Diffusion Models for Text-to-Image Synthesis)

This research-oriented repository aims to conduct a practical evaluation of Stability AI's latent diffusion models in generating images from text prompts. By employing CLIP, Naturalness Image Quality Evaluator (NIQE), Blind/Referenceless Image Spatial Quality Evaluator (BRISQUE), Tenengrad (TENG), and Gradient Magnitude Similarity Deviation (GMSD) scores I analyze and compare image quality, composition, and alignment with textual input. These scores collectively establish a rigorous framework to objectively assess image quality, semantic alignment, and visual composition in the context of text-to-image generation.

About

This research-focused repository seeks to perform a pragmatic assessment of Stability AI's latent diffusion models in their ability to generate images from text prompts. This assessment is achieved by employing a range of scoring metrics and subsequently computing a novel aggregate weighted measure.

Resources

Stars

3 stars

Watchers

1 watching

Forks

Releases

Packages

Contributors

Languages