Blindly Assess Quality of In-the-Wild Videos via Quality-Aware Pre-Training and Motion Perception

• Published in: IEEE transactions on circuits and systems for video technology Vol. 32; no. 9; pp. 5944 - 5958
• Main Authors: Li, Bowen, Zhang, Weixia, Tian, Meng, Zhai, Guangtao, Wang, Xianpei
• Format: Journal Article
• Language: English
• Published: New York IEEE 01.09.2022; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Blind video quality assessment; Distortion; Feature extraction; Image quality; in-the-wild videos; Knowledge management; list-wise ranking loss; Motion perception; Object recognition; Quality assessment; Quality assurance; Three-dimensional displays; Training; transfer learning; Video; Videos
• ISSN: 1051-8215; 1558-2205
• DOI: 10.1109/TCSVT.2022.3164467

Perceptual quality assessment of the videos acquired in the wilds is of vital importance for quality assurance of video services. The inaccessibility of reference videos with pristine quality and the complexity of authentic distortions pose great challenges for this kind of blind video quality assessment (BVQA) task. Although model-based transfer learning is an effective and efficient paradigm for the BVQA task, it remains to be a challenge to explore what and how to bridge the domain shifts for better video representation. In this work, we propose to transfer knowledge from image quality assessment (IQA) databases with authentic distortions and large-scale action recognition with rich motion patterns. We rely on both groups of data to learn the feature extractor and use a mixed list-wise ranking loss function to train the entire model on the target VQA databases. Extensive experiments on six benchmarking databases demonstrate that our method performs very competitively under both individual database and mixed databases training settings. We also verify the rationality of each component of the proposed method and explore a simple ensemble trick for further improvement.