A Zero-Shot Image Classification Method Based on Adversarial Autoencoder Model

Abstract

A zero-shot image classification method based on an adversarial autoencoder model, using an adversarial autoencoder network trained on visible categories to select a network that can best approximate the distribution of visual features and associate visual features with category semantic features The parameters w and v, and then the category semantic feature a of the unseen category t Input it into the network, use the decoder network G to generate visual features, and calculate the Euclidean distance between the generated visual features and the real visual features. Finally, the category with the smallest distance is considered as the predicted category, so as to realize the zero-shot classification task. The present invention is more in line with the characteristics of real data, and at the same time aligns visual features and category semantic features, and can achieve better classification effects in zero-sample tasks.

Description

technical field [0001] The invention relates to a zero-sample classification method. In particular, it concerns a zero-shot classification method based on an adversarial autoencoder model. Background technique [0002] Deep learning has greatly facilitated the development of computer vision, such as object classification, image retrieval, and action recognition, etc. The performance of these tasks is usually evaluated after training with a large amount of labeled data. However, some tasks have only a small portion of training data or even no training data, making traditional classification models perform poorly. In order to improve the classification performance of traditional classification models for categories with little or no data, zero-shot learning has attracted extensive attention. The task of Zero Shot Learning is to classify categories without training data. Humans have the ability to reason, which means that humans can successfully infer the category of unseen...

AI Technical Summary

Problems solved by technology

However, most adversarial networks only focus on generating distributions that approximate real visual features, but ignore the correspondence between visual features and category semantic features, making the generated visual features lack discriminative information to a certain extent.

Images