Presenter(s)

Kriti Sharma, Thomas William Sherk

Comments

Presentation: 11:20-11:40, LTC Studio

Download

Download Project (3.8 MB)

Description

With the rapid advancements in artificial intelligence, text- image generation models such as Generative Adversarial Networks (GANs) and the Contrastive Language-Image Pretraining (CLIP ) models have gained significant attention for their ability to create realistic and visually appealing images from textual descriptions. Evaluating the quality of these generated images, however, remains a challenging task, as these models are complex and some proprietary and the datasets large. To advance the study of artificially generated images, we are introducing a novel dataset, Generative Artificial Image Assessment (GAIA), comprised of images from eight popular text-to-image AI models as well as user rankings from a crowd-sourced annotation. Furthermore, the inclusion of neural net- works/transformer architecture can also obfuscate the underlying decision process that is used to generate the image.

Publication Date

4-17-2024

Project Designation

Graduate Research

Primary Advisor

Van Tam Nguyen

Primary Advisor's Department

Computer Science

Keywords

Stander Symposium, College of Arts and Sciences

Assessment of Generative AI Images

Share

COinS