This website requires JavaScript.

VDPVE: VQA Dataset for Perceptual Video Enhancement

Yixuan GaoYuqin CaoTengchuan Kou ...+4 Guangtao Zhai
Mar 2023
Recently, many video enhancement methods have been proposed to improve videoquality from different aspects such as color, brightness, contrast, andstability. Therefore, how to evaluate the quality of the enhanced video in away consistent with human visual perception is an important research topic.However, most video quality assessment methods mainly calculate video qualityby estimating the distortion degrees of videos from an overall perspective. Fewresearchers have specifically proposed a video quality assessment method forvideo enhancement, and there is also no comprehensive video quality assessmentdataset available in public. Therefore, we construct a Video quality assessmentdataset for Perceptual Video Enhancement (VDPVE) in this paper. The VDPVE has1211 videos with different enhancements, which can be divided into threesub-datasets: the first sub-dataset has 600 videos with color, brightness, andcontrast enhancements; the second sub-dataset has 310 videos with deblurring;and the third sub-dataset has 301 deshaked videos. We invited 21 subjects (20valid subjects) to rate all enhanced videos in the VDPVE. After normalizing andaveraging the subjective opinion scores, the mean opinion score of each videocan be obtained. Furthermore, we split the VDPVE into a training set, avalidation set, and a test set, and verify the performance of severalstate-of-the-art video quality assessment methods on the test set of the VDPVE.