Temporal Information Detection for Video Quality Evaluation

Abstract

Temporal parameters are absolutely necessary in models for video quality evaluation. By detecting the temporal information (TI), it is able to design the proper parameters which can be used in the application of video quality evaluation. Nowadays, the conventional method is to calculate the RMS (root mean square) energy of the temporal gradient. Its workload would enhance greatly when the number of the pixel in image had increased. The paper proposed a new algorithm by calculating the histogram of the temporal gradient. It would increase the efficiency three times more than the conventional one while have the similar performance. The algorithm was so simple and feasible that it could be used for the application of video quality assessment.