Image processing method and device

Abstract

The invention relates to the technical field of deep learning, and provides an image processing method and device. The image processing method comprises the steps of inputting an image to be processedinto a first neural network model, and obtaining a first downsampling position in a second neural network model predicted by the first neural network model; adding a first down-sampling layer at thefirst down-sampling position, wherein the second neural network model after the first down-sampling layer is added is a third neural network model; and inputting the to-be-processed image into the third neural network model to obtain a first image processing result output by the third neural network model. According to the method, the first downsampling position is adaptively predicted according to the to-be-processed image, so that the downsampling operation has higher flexibility, the calculation amount of the second neural network model when executing the image processing task is reduced, and the balance of the second neural network model in the aspects of reducing the calculation amount and ensuring the model precision is favorably realized.

Description

technical field [0001] The present application relates to the technical field of deep learning, in particular, to an image processing method and device. Background technique [0002] With the wide application of neural network models in the field of deep learning, current research focuses on designing deeper networks to improve the accuracy of the model. Taking the image classification task as an example, although the current popular network frameworks (such as ResNet, Inception, etc.) can achieve high classification accuracy, however, a large number of floating-point operations are required for the classification of any image, and its efficiency is relatively low. Contents of the invention [0003] In view of this, the embodiment of the present application provides an image processing method and device, predicting the downsampling position according to the image to be processed, and using the neural network model after adding the downsampling layer at the downsampling pos...

AI Technical Summary

Problems solved by technology

Taking the image classification task as an example, although the current popular network frameworks (such as ResNet, Inception, etc.) can achieve high classification accuracy, however, a large number of floating-point operations are required for the classification of any image, and its efficiency is relatively low.

Images