AGHI-QA: A Subjective-Aligned Dataset and Metric for AI-Generated Human Images
IEEE Transactions on Circuits and Systems for Video Technology · 2026-04 · 36 (4) · 4636-4649
DOI: 10.1109/tcsvt.2025.3639620 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Circuits and Systems for Video Technology, 2026-04. Verfasst von Yunhao Li und 8 weiteren Autor:innen. Die Arbeit wurde bisher 9-mal zitiert (Crossref) und nennt 97 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
The rapid development of text-to-image (T2I) generation approaches has attracted extensive interest in evaluating the quality of generated images, leading to the development of various quality assessment methods for general-purpose T2I outputs. However, existing image quality assessment (IQA) methods are limited to providing global quality scores, failing to deliver fine-grained perceptual evaluations for structurally complex subjects like humans, which is a critical challenge considering the frequent anatomical and textural distortions in AI-generated human images (AGHIs). To address this gap, we introduce AGHI-QA, a large-scale benchmark specifically designed for quality assessment of AGHIs. The dataset comprises $4,000$ images generated from 400 carefully crafted text prompts using 10 state-of-the-art T2I models. We conduct a systematic subjective study to collect multidimensional annotations, including perceptual quality scores, text-image correspondence scores, visible and distorted body part labels. Based on AGHI-QA, we evaluate the strengths and weaknesses of current T2I methods in generating human images from multiple dimensions. Furthermore, we propose AGHI-Assessor, a novel quality metric that integrates the large multimodal model (LMM) with domain-specific human features for precise quality prediction and identification of visible and distorted body parts in AGHIs. Extensive experimental results demonstrate that AGHI-Assessor showcases state-of-the-art performance, significantly outperforming existing IQA methods in multidimensional quality assessment and surpassing leading LMMs in detecting structural distortions in AGHIs.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Li, Y., Wu, S., Sun, W., Zhang, Z., Zhu, Y., Zhang, Z., Duan, H., Min, X., & Zhai, G. (2026). AGHI-QA: A Subjective-Aligned Dataset and Metric for AI-Generated Human Images. IEEE Transactions on Circuits and Systems for Video Technology, 36(4), 4636-4649. https://doi.org/10.1109/tcsvt.2025.3639620
Vancouver
Li Y, Wu S, Sun W, Zhang Z, Zhu Y, Zhang Z, et al. AGHI-QA: A Subjective-Aligned Dataset and Metric for AI-Generated Human Images. IEEE Transactions on Circuits and Systems for Video Technology. 2026;36(4):4636-4649. doi:10.1109/tcsvt.2025.3639620
BibTeX
@article{Li2026,
author = {Li, Yunhao and Wu, Sijing and Sun, Wei and Zhang, Zhichao and Zhu, Yucheng and Zhang, Zicheng and Duan, Huiyu and Min, Xiongkuo and Zhai, Guangtao},
title = {AGHI-QA: A Subjective-Aligned Dataset and Metric for AI-Generated Human Images},
journal = {IEEE Transactions on Circuits and Systems for Video Technology},
year = {2026},
volume = {36},
number = {4},
pages = {4636--4649},
doi = {10.1109/tcsvt.2025.3639620}
}
EUVIMED