Attention-Aware Generalized Mean Pooling for Image Retrieval

TL;DR

This paper introduces an attention-aware generalized mean pooling method for CNN-based image retrieval, significantly improving descriptor quality and retrieval accuracy on challenging benchmarks.

Contribution

It proposes integrating attention mechanisms with GeM pooling to enhance feature relevance in image descriptors, outperforming previous methods.

Findings

Significant performance improvement on ROxford5k and RParis6k benchmarks.

Attention-aware GeM descriptor outperforms state-of-the-art methods.

Notable gains under the 'Hard' evaluation protocol.

Abstract

It has been shown that image descriptors extracted by convolutional neural networks (CNNs) achieve remarkable results for retrieval problems. In this paper, we apply attention mechanism to CNN, which aims at enhancing more relevant features that correspond to important keypoints in the input image. The generated attention-aware features are then aggregated by the previous state-of-the-art generalized mean (GeM) pooling followed by normalization to produce a compact global descriptor, which can be efficiently compared to other image descriptors by the dot product. An extensive comparison of our proposed approach with state-of-the-art methods is performed on the new challenging ROxford5k and RParis6k retrieval benchmarks. Results indicate significant improvement over previous work. In particular, our attention-aware GeM (AGeM) descriptor outperforms state-of-the-art method on ROxford5k under the `Hard' evaluation protocal.