Satisfied user ratio prediction with support vector regression for compressed stereo images
We propose the first method to predict the Satisfied User Ratio (SUR) for compressed stereo images. The method consists of two main steps. First, considering binocular vision properties, we extract three types of features from stereo images: image quality features, monocular visual features, and binocular visual features. Then, we train a Support Vector Regression (SVR) model to learn a mapping function from the feature space to the SUR values. Experimental results on the SIAT-JSSI dataset show excellent prediction accuracy, with a mean absolute SUR error of only 0.08 for H.265 intra coding and only 0.13 for JPEG2000 compression.
Citation : Fan, C., Zhang, Y., Hamzaoui, R., Ziou, D., Jiang, Q. (2020) Satisfied user ratio prediction with support vector regression for compressed stereo images. IEEE International Conference on Multimedia & Expo Workshops (ICMEW), London, July 2020.
Research Institute : Institute of Engineering Sciences (IES)
Peer Reviewed : Yes