摘要 |
An image decoding apparatus obtains the prediction error or a restored image of the encoding object image by inverse processing of the encoding processing of encoded image data encoded by motion compensation predictive coding, in which a restored image is obtained by adding the prediction error and a reference image, a reduced size image of the restored image is generated and stored along with the restored image; wherein, when an image is encoded by an encoding mode that uses reference pixels that include two times or more number of pixels of a region of a predetermined number of pixels, a reference image is obtained by reading out and expanding the stored reduced size image of the restored image, and when an image is encoded using reference pixels of less than two times a number of pixels of the predetermined number of pixels, the reference image is obtained from the stored restored image. |