Mixed-granularity object recognition model training and recognition method and device and storage medium

Abstract

The invention relates to the technical field of the Internet, and discloses a mixed-granularity object recognition model training and recognition method and device and a storage medium. The mixed-granularity object recognition model training method comprises: acquiring sample images, determining category labels of the sample images, and the category labels comprising fine-granularity categories and coarse-granularity categories; performing image category recognition training on an initial deep learning model based on the sample image and the category label of the sample image to obtain a pre-training model; and adjusting the fine-grained branch classification module of the pre-training model by taking the feature difference between the enlarged fine-grained categories as a target to obtaina mixed-grained object recognition model. According to the invention, coarse-grained category identification and fine-grained category identification can be carried out in the same network structure,and the accuracy of fine-grained category identification is improved.

Description

technical field [0001] The present application relates to the field of Internet technologies, and in particular to a method, device and storage medium for training and recognizing a mixed granularity object recognition model. Background technique [0002] In products that implement object recognition, tasks that are often encountered are both coarse-grained and fine-grained recognition. For example, today people like to keep cats, dogs, birds and other pets. People pay attention to the sub-categories of specific animals, because animals of different sub-categories have different habits, preferences and intelligence, such as the border collie under the dog category. , Poodle, Husky, etc. This requires the user to first know which subcategory the animal belongs to, but there are many people who are not familiar with the subcategory of pets. Therefore, in this type of pet identification, it is not only necessary to distinguish coarse-grained categories (cats, dogs, birds, etc....

AI Technical Summary

Problems solved by technology

This kind of brute-force recognition method that does not distinguish between fine-grained and fine-grained objects tends to under-recognize the coarse-grained category of fine-grained objects (such as recognizing cats as a fine-grained category of dogs), while fine-grained classification methods are used to perform fine-grained classification on target features. Computational methods can easily lead to poor results in some types of fine-grained recognition

Images