Method and apparatus for coordination of motion determination over multiple frames
First Claim
1. A method for estimating motion between one reference image and one or more frames in a sequence of two or more frames, each frame consisting of a plurality of pixels of an input image, comprising the steps of:
- estimating motion from the reference image to two or more of the frames to produce motion fields;
transforming the motion fields into a motion matrix, where each row of the motion matrix corresponds to one motion field and each row of the motion matrix contains a vertical and a horizontal component of a motion vector pertaining to each pixel of the reference image; and
performing Principal Component Analysis on the motion matrix to produce a factorization of the motion matrix into the product of a motion score matrix and a motion loading matrix, the motion score matrix comprising a plurality of motion score vectors arranged as column vectors, the motion loading matrix comprising a plurality of motion loading vectors arranged as row vectors, the factorization forming a bilinear model in which one column of said motion score matrix and one corresponding row of said motion loading matrix together comprise a factor of the bilinear model, the number of factors being lower than or equal to the number of said frames;
wherein results from the Principal Component Analysis on the motion matrix are used to influence further estimation of motion from the reference image to one or more of the frames.
1 Assignment
0 Petitions
Accused Products
Abstract
The present invention concerns improved motion estimation in signal records. A method for estimating motion between one reference image and each frame in a sequence of frames, each frame consisting of a plurality of samples of an input signal comprises the steps of: transforming the estimated motion fields into a motion matrix, wherein each row corresponds to one frame, and each row contains each component of motion vector for each element of the reference image; performing a Principal Component Analysis of the motion matrix, thereby obtaining a motion score matrix consisting of a plurality of column vectors called motion score vectors and a motion loading matrix consisting of a plurality of row vectors called motion loading vectors, such that each motion score vector corresponds to one element for each frame, such that each element of each motion loading vector corresponds to one element of the reference image, such that one column of said motion score matrix and one motion loading vector together constitute a factor, and such that the number of factors is lower than or equal to the number of said frames; wherein the results from the Principal Component Analysis on the motion matrix are used to influence further estimation of motion from the reference image to one or more of the frames.
137 Citations
43 Claims
-
1. A method for estimating motion between one reference image and one or more frames in a sequence of two or more frames, each frame consisting of a plurality of pixels of an input image, comprising the steps of:
-
estimating motion from the reference image to two or more of the frames to produce motion fields; transforming the motion fields into a motion matrix, where each row of the motion matrix corresponds to one motion field and each row of the motion matrix contains a vertical and a horizontal component of a motion vector pertaining to each pixel of the reference image; and performing Principal Component Analysis on the motion matrix to produce a factorization of the motion matrix into the product of a motion score matrix and a motion loading matrix, the motion score matrix comprising a plurality of motion score vectors arranged as column vectors, the motion loading matrix comprising a plurality of motion loading vectors arranged as row vectors, the factorization forming a bilinear model in which one column of said motion score matrix and one corresponding row of said motion loading matrix together comprise a factor of the bilinear model, the number of factors being lower than or equal to the number of said frames; wherein results from the Principal Component Analysis on the motion matrix are used to influence further estimation of motion from the reference image to one or more of the frames. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41)
-
-
42. An apparatus for estimating motion between one reference image and one or more frames in a sequence of two or more frames, each frame consisting of a plurality of pixels of an input image, said apparatus comprising:
-
means for estimating motion from the reference image to two or more of the frames to produce motion fields; means for transforming the motion fields into a motion matrix, where each row of the motion matrix corresponds to one motion field and each row of the motion matrix contains a vertical and a horizontal component of a motion vector pertaining to each pixel of the reference image; and means for performing Principal Component Analysis on the motion matrix to produce a factorization of the motion matrix into the product of a motion score matrix and a motion loading matrix, the motion score matrix comprising a plurality of motion score vectors arranged as column vectors, the motion loading matrix comprising a plurality of motion loading vectors arranged as row vectors, the factorization forming a bilinear model in which one column of said motion score matrix and one corresponding row of said motion loading matrix together comprise a factor of the bilinear model, the number of factors being lower than or equal to the number of said frames; wherein results from the Principal Component Analysis on the motion matrix are used to influence further estimation of motion from the reference image to one or more of the frames.
-
-
43. A data structure stored in a computer readable medium and suitable for representing motion between one reference image and each frame in a sequence of frames, said frames consisting of a plurality of data samples arranged in a spatial pattern, said data structure comprising:
-
(1) a plurality of motion patterns represented as loading vectors, each element of each loading vector corresponding to one element of said reference image; and (2) a plurality of motion score vectors, each motion score vector corresponding to one of said frames, each motion score vector consisting of the same number of elements as the number of loading vectors, each motion score element of each motion score vector representing how much the corresponding loading vector should contribute to the total motion for said one frame.
-
Specification