摘要 |
PROBLEM TO BE SOLVED: To estimate accurate and uniform subjective quality with respect to an arbitrary video. SOLUTION: An alignment information generating portion (21) inputs a reference video signal (RI) which is an undeteriorated video signal and a deteriorated video signal (PI) which is produced by the deterioration of the reference video signal, and generates alignment information related to a temporal deviation and a spatial deviation between a reference video frame and a deteriorated video frame. The temporal deviation and the spatial deviation are eliminated, on the basis of the alignment information; additionally, a spatial feature amount calculating device (22) calculates each spatial feature amount on the basis of spatial information, showing video states of the reference video frame and the deteriorated video frame, and a temporal feature amount calculator (23) calculates each temporal feature amount, on the basis of temporal information showing video changes between each frame with regard to the reference video frame and the deteriorated video frame. A subjective quality estimating portion (24) estimates the subjective quality of the deteriorated video signal, on the basis of the spatial feature amount and the temporal feature amount. COPYRIGHT: (C)2009,JPO&INPIT
|