Large-Scale Image Retrieval with Attentive Deep Local Features

TL;DR

This paper introduces DELF, an attentive local feature descriptor based on CNNs, designed for large-scale image retrieval, with a new dataset and demonstrated superior performance over existing methods.

Contribution

The paper presents DELF, a novel CNN-based local feature descriptor with an attention mechanism for keypoint selection, optimized for large-scale image retrieval tasks.

Findings

DELF outperforms state-of-the-art descriptors in large-scale retrieval

Introduces the Google-Landmarks dataset for evaluation

Provides a robust confidence scoring mechanism for false positive rejection

Abstract

We propose an attentive local feature descriptor suitable for large-scale image retrieval, referred to as DELF (DEep Local Feature). The new feature is based on convolutional neural networks, which are trained only with image-level annotations on a landmark image dataset. To identify semantically useful local features for image retrieval, we also propose an attention mechanism for keypoint selection, which shares most network layers with the descriptor. This framework can be used for image retrieval as a drop-in replacement for other keypoint detectors and descriptors, enabling more accurate feature matching and geometric verification. Our system produces reliable confidence scores to reject false positives---in particular, it is robust against queries that have no correct match in the database. To evaluate the proposed descriptor, we introduce a new large-scale dataset, referred to as Google-Landmarks dataset, which involves challenges in both database and query such as background clutter, partial occlusion, multiple landmarks, objects in variable scales, etc. We show that DELF outperforms the state-of-the-art global and local descriptors in the large-scale setting by significant margins. Code and dataset can be found at the project webpage: https://github.com/tensorflow/models/tree/master/research/delf .