Joint Hierarchical Category Structure Learning and Large-Scale Image Classification

TL;DR

This paper introduces a scalable image classification approach that learns hierarchical inter-class structures to improve efficiency and accuracy in large-scale multi-class problems, demonstrated on benchmark datasets.

Contribution

A novel method for learning hierarchical visual structures using spectral clustering, enabling efficient and more accurate large-scale image classification.

Findings

Outperforms existing visual tree-based methods in hierarchy quality

Achieves significantly better classification accuracy on benchmark datasets

Provides a fast algorithm for inter-class similarity computation

Abstract

We investigate the scalable image classification problem with a large number of categories. Hierarchical visual data structures are helpful for improving the efficiency and performance of large-scale multi-class classification. We propose a novel image classification method based on learning hierarchical inter-class structures. Specifically, we first design a fast algorithm to compute the similarity metric between categories, based on which a visual tree is constructed by hierarchical spectral clustering. Using the learned visual tree, a test sample label is efficiently predicted by searching for the best path over the entire tree. The proposed method is extensively evaluated on the ILSVRC2010 and Caltech 256 benchmark datasets. Experimental results show that our method obtains significantly better category hierarchies than other state-of-the-art visual tree-based methods and, therefore, much more accurate classification.