Abstract: A reverse filter operates for adding noise n(x,y) to an output of a deteriorated model of a blurring function H(x,y) to output an observed model g(x,y). The blurring function inputs a true picture f(x,y) to output a deteriorated picture. The reverse filter recursively optimizes the blurring function H(x,y) so that the input picture signal will be coincident with the observed picture. In this manner, the reverse filter extracts a true picture signal. A corresponding point is estimated, based on a fluency theory, on the true input picture signal freed of noise contained in it by the reverse filter (20). The motion information of a picture is expressed in the form of a function. A plurality of signal spaces is selected by an encoder for compression (30) for the input picture signal. The picture information is expressed by a function from one selected signal space to another.

Abstract: For each of a plural number of pixels in a reference frame, a corresponding point estimation unit (2) estimates a corresponding point in each of a plural number of picture frames differing in time. A first gray scale value generation unit (3) finds the gray scale value of each pixel from the gray scale values of the neighboring pixels for each corresponding point in each picture frame estimated. A second gray scale value generation unit (4) approximates the gray level of a locus of corresponding points, by a fluency function, from the gray scale values of the corresponding points at each picture frame estimated to find each gray scale value of the corresponding point in the picture frame for interpolation from the above function. From the gray scale value of each corresponding point in the frame for interpolation, a third gray scale value generation unit (5) generates a gray scale value of each pixel in the picture frame for interpolation.