| CPC G06T 5/005 (2013.01) [G06N 3/08 (2013.01); G06T 5/50 (2013.01); G06T 11/60 (2013.01); G06T 2207/20081 (2013.01)] | 20 Claims |
|
1. An image processing method, performed by a computing device, the method comprising:
receiving an input image;
determining a context feature of the input image;
determining a first feature set and a second feature set according to the context feature and based on a size of a target image and a location of the input image in the target image;
adjusting the second feature set according to a first feature statistic of the first feature set, to obtain an adjusted second feature set; and
generating the target image based on the adjusted second feature set and the first feature set;
wherein:
the image processing method is implemented by using a deep neural network, the deep neural network being trained using the following operations:
determining a sample image from a training sample set, and randomly determining a partial image in the sample image as an input of the deep neural network;
processing the partial image by using the deep neural network, and outputting a target image based on the partial image; and
adjusting a value of the deep neural network, to minimize a loss between the target image and the sample image, the loss being determined by a pixel difference between the sample image and the target image based on a matrix of a real sample image, an output of the deep neural network, the input image, a size of an edge, a parameter of the deep neural network, and a weight matrix.
|