Image processing apparatus and operation method thereof
An image processing apparatus for performing an image by using one or more neural networks may include a memory storing one or more instructions and at least one processor configured to execute the one or more instructions to obtain classification information of a first image and first feature information of the first image, generate a first feature image for the first image by performing first image processing on the classification information and the first feature information, obtain second feature information by performing second image processing on the classification information and the first feature information, obtain fourth feature information by performing third image processing on third feature information extracted during the first image processing, generate a second feature image for the first image, based on the second feature information and the fourth feature information, and generate a second image based on the first feature image and the second feature image.
1 . An image processing apparatus comprising:
a memory storing at least one instruction; and
at least one processor configured to execute the at least one instruction to:
obtain classification information of a first image and first feature information of the first image,
generate a first feature image for the first image by performing first image processing on the classification information and the first feature information,
obtain second feature information by performing second image processing on the classification information and the first feature information,
obtain fourth feature information by performing third image processing on third feature information extracted during the first image processing,
generate a second feature image for the first image, based on the second feature information and the fourth feature information, and
generate a second image based on the first feature image and the second feature image.
2 . The image processing apparatus of claim 1 , wherein the first feature image comprises features of non-high frequency components in the first image, and
wherein the second feature image comprises features of high frequency components in the first image.
3 . The image processing apparatus of claim 1 , wherein a quality of the second image is higher than a quality of the first image.
4 . The image processing apparatus of claim 1 , wherein the at least one processor is further configured to execute the at least one instruction to obtain the classification information and the first feature information by using at least one convolutional neural network.
5 . The image processing apparatus of claim 1 , wherein the first image processing comprises upsampling the first feature information, and
wherein the first image, the first feature image, and the second feature image have a same size.
6 . The image processing apparatus of claim 1 , wherein the second image processing is performed by a multi-layer perceptron (MLP) module comprising at least one fully connected layer,
wherein the at least one processor is further configured to execute the at least one instruction to perform a multiplication operation between an input data fed to a fully connected layer and a weight matrix in the fully connected layer and an addition operation with biases in the fully connected layer.
7 . The image processing apparatus of claim 1 , wherein the at least one processor is further configured to execute the at least one instruction to:
obtain a sub-feature information by performing downscaling and upscaling on the third feature information;
obtain a difference information between the sub-feature information and the third feature information; and
generate the fourth feature information by performing a convolution operation between the difference information and a pre-trained weight.
8 . The image processing apparatus of claim 1 , wherein the at least one processor is further configured to execute the at least one instruction to:
obtain fifth feature information by performing a first operation on the second feature information; and
generate the second feature image by performing a second operation on the fifth feature information and the fourth feature information.
9 . The image processing apparatus of claim 8 , wherein the first operation comprises an adaptive instance normalization (AdaIn) operation, and
wherein the second operation comprises a spatial feature transform (SFT) operation.
10 . A method of operating an image processing apparatus comprising:
obtaining classification information of a first image and first feature information of the first image;
generating a first feature image for the first image by performing first image processing on the classification information and the first feature information;
obtaining second feature information by performing second image processing on the classification information and the first feature information;
obtaining fourth feature information by performing third image processing on third feature information extracted during the first image processing;
generating a second feature image for the first image, based on the second feature information and the fourth feature information; and
generating a second image based on the first feature image and the second feature image.
11 . The method of claim 10 , wherein the first feature image comprises features of non-high frequency components in the first image, and
wherein the second feature image comprises features of high frequency components in the first image.
12 . The method of claim 10 , wherein a quality of the second image is higher than a quality of the first image.
13 . The method of claim 10 , wherein the obtaining the classification information and the first feature information comprises:
obtaining the classification information and the first feature information of the first image by using at least one convolutional neural network.
14 . The method of claim 10 , wherein the first image processing comprises upsampling of the first feature information, and
wherein the first image, the first feature image, and the second feature image have a same size.
15 . The method of claim 10 , wherein the second image processing is performed by a multi-layer perceptron (MLP) module comprising at least one fully connected layer, and
wherein the obtaining the second feature information by performing the second image processing on the classification information and the first feature information comprises:
performing a multiplication operation between input data fed to a fully connected layer and a weight matrix in the fully connected layer and an addition operation with biases in the fully connected layer.
16 . The method of claim 10 , wherein the obtaining the fourth feature information by performing the third image processing on the third feature information extracted during the first image processing comprises:
obtaining sub-feature information by performing downscaling and upscaling on the third feature information;
obtaining difference information between the third feature information and the sub-feature information; and
generating the fourth feature information by performing a convolution operation between the difference information and a pre-trained weight.
17 . The method of claim 10 , wherein the generating the second feature image for the first image based on the second feature information and the fourth feature information comprises:
obtaining fifth feature information by performing a first operation on the second feature information; and
generating the second feature image by performing a second operation on the fifth feature information and the fourth feature information.
18 . The method of claim 17 , wherein the first operation comprises an adaptive instance normalization (AdaIn) operation, and
wherein the second operation comprises a spatial feature transform (SFT) operation.
19 . The method of claim 10 , wherein the third feature information comprises a plurality of pieces of intermediate data that is output by a process of generating the first feature image.
20 . A non-transitory computer-readable recording medium storing computer readable program code or instructions which are executable by a processor to perform a method of image processing, the method comprising:
obtaining classification information of a first image and first feature information of the first image;
generating a first feature image for the first image by performing first image processing on the classification information and the first feature information;
obtaining second feature information by performing second image processing on the classification information and the first feature information;
obtaining fourth feature information by performing third image processing on third feature information extracted during the first image processing;
generating a second feature image for the first image, based on the second feature information and the fourth feature information; and
generating a second image based on the first feature image and the second feature image.