Comparison Knowledge Translation for Generalizable Image Classification

TL;DR

This paper introduces a novel framework that emulates human recognition to improve image classification, especially on unseen categories, by translating comparison knowledge from labeled to novel categories using a specialized neural network.

Contribution

It proposes the Comparison Knowledge Translation (CKT) task and the CCT-Net model, which enhances generalization to unseen categories through comparison-based learning and adversarial training.

Findings

CCT-Net achieves state-of-the-art performance on target categories.

The framework demonstrates strong generalization to unseen categories.

Experimental results validate the effectiveness of comparison knowledge translation.

Abstract

Deep learning has recently achieved remarkable performance in image classification tasks, which depends heavily on massive annotation. However, the classification mechanism of existing deep learning models seems to contrast to humans' recognition mechanism. With only a glance at an image of the object even unknown type, humans can quickly and precisely find other same category objects from massive images, which benefits from daily recognition of various objects. In this paper, we attempt to build a generalizable framework that emulates the humans' recognition mechanism in the image classification task, hoping to improve the classification performance on unseen categories with the support of annotations of other categories. Specifically, we investigate a new task termed Comparison Knowledge Translation (CKT). Given a set of fully labeled categories, CKT aims to translate the comparison knowledge learned from the labeled categories to a set of novel categories. To this end, we put forward a Comparison Classification Translation Network (CCT-Net), which comprises a comparison classifier and a matching discriminator. The comparison classifier is devised to classify whether two images belong to the same category or not, while the matching discriminator works together in an adversarial manner to ensure whether classified results match the truth. Exhaustive experiments show that CCT-Net achieves surprising generalization ability on unseen categories and SOTA performance on target categories.