MULTIVIEW IMAGE ENCODING METHOD, MULTIVIEW IMAGE DECODING METHOD, MULTIVIEW IMAGE ENCODING DEVICE, MULTIVIEW IMAGE DECODING DEVICE, AND PROGRAMS OF SAME

摘要

This multiview image encoding method includes the following steps: a view synthesis image production step that produces a view synthesis image for the view that is subject to encoding, using a reference view image; a prediction image candidate production step that produces, for each region that is subject to encoding, a plurality of intra prediction image candidates from the decoded images of regions near the region subject to encoding that have already been encoded; a prediction image candidate assessment step that determines the assessment values of the intra prediction image candidates using the view synthesis image; a prediction image production step that produces an intra prediction image from the intra prediction image candidates on the basis of the assessment values; and an image signal encoding step that performs predictive encoding of the image that is subject to encoding from the region that is subject to encoding using the intra prediction image.