Deep Image Retrieval: Learning global representations for image search

TL;DR

This paper introduces a deep learning method for image retrieval that learns to generate compact global image descriptors by training on a large, cleaned dataset, outperforming previous global and local descriptor methods.

Contribution

It presents a novel deep architecture trained specifically for image retrieval, using a ranking framework and region proposal network to optimize global image representations.

Findings

Outperforms previous global descriptor methods on standard datasets.

Surpasses many local descriptor-based approaches in accuracy.

Efficient single-pass architecture for global image representation.

Abstract

We propose a novel approach for instance-level image retrieval. It produces a global and compact fixed-length representation for each image by aggregating many region-wise descriptors. In contrast to previous works employing pre-trained deep networks as a black box to produce features, our method leverages a deep architecture trained for the specific task of image retrieval. Our contribution is twofold: (i) we leverage a ranking framework to learn convolution and projection weights that are used to build the region features; and (ii) we employ a region proposal network to learn which regions should be pooled to form the final global descriptor. We show that using clean training data is key to the success of our approach. To that aim, we use a large scale but noisy landmark dataset and develop an automatic cleaning approach. The proposed architecture produces a global image representation in a single forward pass. Our approach significantly outperforms previous approaches based on global descriptors on standard datasets. It even surpasses most prior works based on costly local descriptor indexing and spatial verification. Additional material is available at www.xrce.xerox.com/Deep-Image-Retrieval.