User-generated Video Quality Assessment: A Subjective and Objective Study

TL;DR

This paper investigates the quality of user-generated videos through a comprehensive study involving a new database and a deep learning-based assessment algorithm, aiming to improve perceptual quality evaluation for better video processing.

Contribution

It introduces a new UGC video quality assessment database and a deep neural network-based algorithm for objective quality evaluation considering corrupted references.

Findings

Proposed method outperforms existing quality assessment algorithms.

Constructed a comprehensive UGC video quality database with subjective scores.

Insights provided for perceptual UGC video coding and processing.

Abstract

Recently, we have observed an exponential increase of user-generated content (UGC) videos. The distinguished characteristic of UGC videos originates from the video production and delivery chain, as they are usually acquired and processed by non-professional users before uploading to the hosting platforms for sharing. As such, these videos usually undergo multiple distortion stages that may affect visual quality before ultimately being viewed. Inspired by the increasing consensus that the optimization of the video coding and processing shall be fully driven by the perceptual quality, in this paper, we propose to study the quality of the UGC videos from both objective and subjective perspectives. We first construct a UGC video quality assessment (VQA) database, aiming to provide useful guidance for the UGC video coding and processing in the hosting platform. The database contains source UGC videos uploaded to the platform and their transcoded versions that are ultimately enjoyed by end-users, along with their subjective scores. Furthermore, we develop an objective quality assessment algorithm that automatically evaluates the quality of the transcoded videos based on the corrupted reference, which is in accordance with the application scenarios of UGC video sharing in the hosting platforms. The information from the corrupted reference is well leveraged and the quality is predicted based on the inferred quality maps with deep neural networks (DNN). Experimental results show that the proposed method yields superior performance. Both subjective and objective evaluations of the UGC videos also shed lights on the design of perceptual UGC video coding.