Video quality assessment method and apparatus

A computing apparatus comprising:

• a memory configured to store one or more instructions;
• and
• a processor configured to execute the one or more instructions stored in the memory to:
• obtain a reference image;
• obtain a distorted image generated from the reference image;
• obtain an objective quality assessment score of the distorted image, that is indicative of a quality of the distorted image as assessed by an algorithm, by using the reference image and the distorted image;
• obtain a subjective quality assessment score, that is indicative of a quality of the distorted image as assessed by a human, corresponding to the objective quality assessment score;
• and
• train a neural network, by using the distorted image and the subjective quality assessment score as a training data set.

An image display apparatus comprising:

• a display;
• a memory configured to store one or more instructions;
• and
• a processor configured to execute the one or more instructions stored in the memory to:
• input a frame to a trained neural network,
• obtain a quality score with respect to the frame from the trained neural network,
• post-process the frame based on the quality score of the frame, and
• provide the post-processed frame for output via the display,
• wherein the trained neural network is trained by obtaining a reference image;
• obtaining a distorted image generated from the reference image;
• obtaining an objective quality assessment score of the distorted image that identifies a quality of the distorted image as assessed by an algorithm, by using the reference image and the distorted image;
• obtaining a subjective quality assessment score corresponding to the objective quality assessment score;
• and training by using the distorted image and the subjective quality assessment score as a training data set.