Encoding method, encoding apparatus, and program
An encoding method is performed by an encoding apparatus, and the encoding method includes a step in which an image of an object captured by a camera in a plurality of frames of a moving image is analyzed and it is determined whether camera parameters relating to an intrinsic matrix of the camera are invariable, a step in which, when the camera parameters are invariable, six parameters representing destinations of three points for a projective transformation unit of the image of the object are generated, and a homography matrix is generated on the basis of the six parameters, and a step in which projective transformation is performed on the projective transformation unit of the image of the object using the homography matrix, and a prediction signal of the image of the object is generated on the basis of the result of the projective transformation.
1. An encoding method performed by an encoding apparatus, comprising:
steps in which, on the basis of an image of an object captured by a camera in a plurality of frames of a moving image, it is determined whether camera parameters relating to an intrinsic matrix of the camera are the same in the plurality of frames, and the camera parameters are output to a predetermined apparatus when the camera parameters are the same;
steps in which, when the camera parameters are the same, six parameters representing destinations of three points defined in the image of the object are generated by a six-dimensional search which updates a third column of a matrix representing a space rotation amount when the camera moves between two positions, a homography matrix is generated on the basis of the six parameters, and the six parameters are output to the predetermined apparatus;
a step in which projective transformation using the homography matrix is performed on a decoded image of the moving image;
a step in which a prediction residual signal which is a difference between the result of the projective transformation and the frame is generated; and
a step in which the prediction residual signal is encoded.
2. The encoding method according to claim 1 ,
wherein the object has a flat exterior shape.
3. An encoding apparatus, comprising:
an analyzer configured to, on the basis of an image of an object captured by a camera in a plurality of frames of a moving image, determine whether camera parameters relating to an intrinsic matrix of the camera are the same in the plurality of frames, and output the camera parameters to a predetermined apparatus when the camera parameters are the same;
a matrix generator configured to, when the camera parameters are the same, generate six parameters representing destinations of three points defined in the image of the object by a six-dimensional search which updates a third column of a matrix representing a space rotation amount when the camera moves between two positions, generate a homography matrix on the basis of the six parameters, and output the six parameters to the predetermined apparatus;
a projective transformer configured to perform projective transformation using the homography matrix on a decoded image of the moving image;
a subtracter configured to generate a prediction residual signal which is a difference between the result of the projective transformation and the frame; and
an encoder configured to encode the prediction residual signal,
wherein each of the analyzer, the matrix generator, the projective transformer, the subtracter and the encoder is implemented by:
i) computer executable instructions executed by the at least one processor,
ii) at least one circuitry or
iii) a combination of computer executable instructions executed by at least one processor and at least one circuitry.
4. The encoding apparatus according to claim 3 ,
wherein the object has a flat exterior shape.
5. A non-transitory computer readable medium storing a program causing a computer to function as the encoding apparatus according to claim 3 .