JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering

Accepted manuscript available online (unedited version)

Image generation evaluation: a comprehensive survey of human and automatic evaluations

Author(s): Qi LIU¹, Shuanglin YANG², Zejian LI¹, Lefan HOU³, Chenye MENG³, Ying ZHANG¹, Lingyun SUN³
Affiliation(s): ¹School of Software Technology, Zhejiang University, Ningbo 315100, China; more
Corresponding email(s): zejianlee@zju.edu.cn
Key Words: Image generation evaluation; Human evaluation; Automatic evaluation; Evaluation protocols; Evaluation aspects

Share this article to： More <<< Previous Paper \|Next Paper >>>

Qi LIU¹, Shuanglin YANG², Zejian LI¹, Lefan HOU³, Chenye MENG³,Ying ZHANG¹, Lingyun SUN³. Image generation evaluation: a comprehensive survey of human and automatic evaluations[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.2400904

@article{title="Image generation evaluation: a comprehensive survey of human and automatic evaluations",
author="Qi LIU¹, Shuanglin YANG², Zejian LI¹, Lefan HOU³, Chenye MENG³,Ying ZHANG¹, Lingyun SUN³",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.2400904"
}

%0 Journal Article
%T Image generation evaluation: a comprehensive survey of human and automatic evaluations
%A Qi LIU¹
%A Shuanglin YANG²
%A Zejian LI¹
%A Lefan HOU³
%A Chenye MENG³
%A Ying ZHANG¹
%A Lingyun SUN³
%J Frontiers of Information Technology & Electronic Engineering
%P
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.2400904"

TY - JOUR
T1 - Image generation evaluation: a comprehensive survey of human and automatic evaluations
A1 - Qi LIU¹
A1 - Shuanglin YANG²
A1 - Zejian LI¹
A1 - Lefan HOU³
A1 - Chenye MENG³
A1 - Ying ZHANG¹
A1 - Lingyun SUN³
J0 - Frontiers of Information Technology & Electronic Engineering
SP -
EP -
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.2400904"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Image generation models have made remarkable progress, and image evaluation is crucial for explaining and driving the development of these models. Previous studies have extensively explored human and automatic evaluations of image generation. Herein, these studies are comprehensively surveyed, specifically for two main parts: evaluation protocols and evaluation methods. First, 10 image generation tasks are summarized by focusing on their differences in evaluation aspects. Based on this, a novel protocol is proposed to cover human and automatic evaluation aspects required for various image generation tasks. Second, the review of automatic evaluation methods in the past five years is highlighted. To our knowledge, this paper presents the first comprehensive summary of human evaluation, encompassing evaluation methods, tools, details, and data analysis methods. Finally, this the challenges and potential directions for image generation evaluation are discussed. We hope that this survey will help researchers to develop a systematic understanding of image generation evaluation, stay updated with the latest advancements in the field, and encourage further research.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference