A Semantic Indexing Structure for Image Retrieval

TL;DR

This paper introduces a Semantic Indexing Structure (SIS) for large-scale image retrieval that leverages semantic categories for database partitioning, overcoming high-dimensionality and variable feature sizes, and demonstrating superior performance.

Contribution

The paper proposes a novel semantic classification-based indexing method that integrates with feature extractors regardless of feature dimension and size, improving retrieval efficiency.

Findings

SIS effectively handles high-dimensional and variable-sized features.

Normalized partition number to five yields optimal performance.

SIS outperforms existing state-of-the-art models in tests.

Abstract

In large-scale image retrieval, many indexing methods have been proposed to narrow down the searching scope of retrieval. The features extracted from images usually are of high dimensions or unfixed sizes due to the existence of key points. Most of existing index structures suffer from the dimension curse, the unfixed feature size and/or the loss of semantic similarity. In this paper a new classification-based indexing structure, called Semantic Indexing Structure (SIS), is proposed, in which we utilize the semantic categories rather than clustering centers to create database partitions, such that the proposed index SIS can be combined with feature extractors without the restriction of dimensions. Besides, it is observed that the size of each semantic partition is positively correlated with the semantic distribution of database. Along this way, we found that when the partition number is normalized to five, the proposed algorithm performed very well in all the tests. Compared with state-of-the-art models, SIS achieves outstanding performance.