Challenge Introduction

In recent years, image style transfer and AI painting have become high-profile research areas. Users can easily generate artworks of numerous styles from websites and online applications, which has led to the explosion of artistic images online and the drastic increase in demand for automatically evaluating artwork aesthetics.  However, existing datasets like AVA and AADB contain few artworks and most of the state-of-the-art image aesthetic assessment methods are targeted towards real-world photos. In this challenge, we provide a dataset which contains 60337 artistic images with aesthetic score annotations for each image. The competition tends to inspire researchers to rethink AIAA and stimulate research with a broader perspective.

Challenge Organization

Challenge Co-Chairs

  • Ran Yi, Shanghai Jiao Tong University, China

Organizing Team

  • Ran Yi, Shanghai Jiao Tong University, China
  • Haoyuan Tian, Shanghai Jiao Tong University, China
  • Hu Teng, Shanghai Jiao Tong University, China
  • Yu-Kun Lai, Cardiff University, United Kingdom
  • Paul L. Rosin, Cardiff University, United Kingdom

Challenge Webpage

CGI-AIAA2023

Challenge Contact