Leveraging Implicit Spatial Information in Global Features for Image Retrieval

TL;DR

This paper introduces ISTA, a novel image retrieval method that incorporates implicit spatial information into global features, significantly improving retrieval accuracy on standard datasets.

Contribution

The paper presents a new tensor-based aggregation method that preserves relative spatial information in global image features, advancing state-of-the-art performance.

Findings

Achieves top performance on Holidays, Oxford5k, and Paris6k datasets.

Outperforms existing global feature aggregation methods.

Demonstrates the effectiveness of spatial information integration in image retrieval.

Abstract

Most image retrieval methods use global features that aggregate local distinctive patterns into a single representation. However, the aggregation process destroys the relative spatial information by considering orderless sets of local descriptors. We propose to integrate relative spatial information into the aggregation process by taking into account co-occurrences of local patterns in a tensor framework. The resulting signature called Improved Spatial Tensor Aggregation (ISTA) is able to reach state of the art performances on well known datasets such as Holidays, Oxford5k and Paris6k.