Image retrieval with hierarchical matching pursuit

TL;DR

This paper introduces a hierarchical sparse coding method for image retrieval that effectively fuses multi-scale cues, leading to improved performance with compact codes.

Contribution

The paper proposes a novel hierarchical sparse coding architecture that captures multi-scale features for image retrieval, enhancing discriminative power over fixed-scale methods.

Findings

Achieves excellent retrieval performance on the Holidays dataset.

Uses small code length for efficient retrieval.

Effectively fuses multi-scale cues through hierarchical sparse coding.

Abstract

A novel representation of images for image retrieval is introduced in this paper, by using a new type of feature with remarkable discriminative power. Despite the multi-scale nature of objects, most existing models perform feature extraction on a fixed scale, which will inevitably degrade the performance of the whole system. Motivated by this, we introduce a hierarchical sparse coding architecture for image retrieval to explore multi-scale cues. Sparse codes extracted on lower layers are transmitted to higher layers recursively. With this mechanism, cues from different scales are fused. Experiments on the Holidays dataset show that the proposed method achieves an excellent retrieval performance with a small code length.