Zero-sample classifying method based on class transfer

Abstract

A zero-sample classifying method based on class transfer comprises the steps of acquiring a vision characteristic of C kinds of training samples, a class semantic characteristic of the training sampleand a true label matrix; calculating a semantic similarity matrix by means of cosine similarity or Gaussian similarity through the class semantic characteristic; calculating a diagonal matrix of a class semantic similarity matrix; calling a Sylvester equation in an MATLAB toolset for obtaining a mapping matrix; inputting the vision characteristic of the training sample, the corresponding class semantic characteristic and the true label matrix into a target function, continuously adjusting the value of a model regularization parameter, calculating the least value of the target function, and finishing model training; and in a testing period, inputting the vision characteristic of the testing sample and the corresponding semantic characteristic, calculating scores of the classes, and determining the class with highest score as the predicated class of the testing sample. The zero-sample classifying method based on class transfer has advantages of sufficiently digging the semantic relationbetween different classes, realizing knowledge transfer between a known class classifier and an unknown class classifier, and realizing high convenience in application in image classification.

Description

technical field [0001] The invention relates to a zero-sample classification method. In particular, it concerns a class-transfer-based zero-shot classification method that enables knowledge transfer from known to unknown classes. Background technique [0002] Image classification technology plays an important role in obtaining image information quickly and accurately. Traditional image classification needs to provide labeled training samples for all categories during the training phase. On the one hand, with the development of information and multimedia technology, the number and types of images have increased greatly, and the workload of manual labeling is huge, and the efficiency is low. The requirements are high; on the other hand, for some rare categories, training samples are often difficult to obtain. Therefore, it is unrealistic to provide manually labeled training samples for all target categories. How to solve the classification problem of target categories withou...

AI Technical Summary

Problems solved by technology

Traditional image classification needs to provide labeled training samples for all categories during the training phase. On the one hand, with the development of information multimedia technology, the number and types of images have increased greatly, and manual labeling has a huge workload and low efficiency. High requirements; on the other hand, for some rare categories, training samples are often difficult to obtain

Images