AIS 2024 Challenge on Video Quality Assessment of User-Generated Content: Methods and Results

This paper reviews the AIS 2024 Video Quality Assessment (VQA) Challenge,focused on User-Generated Content (UGC). The aim of this challenge is to gatherdeep learning-based methods capable of estimating the perceptual quality of UGCvideos. The user-generated videos from the YouTube UGC Dataset include diversecontent (sports, games, lyrics, anime, etc.), quality and resolutions. Theproposed methods must process 30 FHD frames under 1 second. In the challenge, atotal of 102 participants registered, and 15 submitted code and models. Theperformance of the top-5 submissions is reviewed and provided here as a surveyof diverse deep models for efficient video quality assessment of user-generatedcontent.