Characterizing Photorealism and Artifacts in Diffusion Model-Generated Images

Negar Kamali, Karyn Nakamura, Aakriti Kumar, Angelos Chatzimparmpas, Jessica Hullman, Matthew Groh

Research output: Chapter in Book/Report/Conference proceedingConference contributionAcademicpeer-review

Abstract

Diffusion model-generated images can appear indistinguishable from authentic photographs, but these images often contain artifacts and implausibilities that reveal their AI-generated provenance. Given the challenge to public trust in media posed by photorealistic AI-generated images, we conducted a large-scale experiment measuring human detection accuracy on 450 diffusion-model generated images and 149 real images. Based on collecting 749,828 observations and 34,675 comments from 50,444 participants, we find that scene complexity of an image, artifact types within an image, display time of an image, and human curation of AI-generated images all play significant roles in how accurately people distinguish real from AI-generated images. Additionally, we propose a taxonomy characterizing artifacts often appearing in images generated by diffusion models. Our empirical observations and taxonomy offer nuanced insights into the capabilities and limitations of diffusion models to generate photorealistic images in 2024.

Original languageEnglish
Title of host publicationCHI 2025 - Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems
PublisherAssociation for Computing Machinery
ISBN (Electronic)9798400713941
DOIs
Publication statusPublished - 26 Apr 2025
Event2025 CHI Conference on Human Factors in Computing Systems, CHI 2025 - Yokohama, Japan
Duration: 26 Apr 20251 May 2025

Publication series

NameConference on Human Factors in Computing Systems - Proceedings

Conference

Conference2025 CHI Conference on Human Factors in Computing Systems, CHI 2025
Country/TerritoryJapan
CityYokohama
Period26/04/251/05/25

Bibliographical note

Publisher Copyright:
© 2025 Copyright held by the owner/author(s).

Keywords

  • deepfakes
  • diffusion models
  • generative AI
  • misinformation
  • photorealism
  • synthetic media

Fingerprint

Dive into the research topics of 'Characterizing Photorealism and Artifacts in Diffusion Model-Generated Images'. Together they form a unique fingerprint.

Cite this