Real-time Local Feature with Global Visual Information Enhancement

TL;DR

This paper introduces a CNN-based local feature extraction method enhanced with global visual information and optimized via deep reinforcement learning, achieving robustness and real-time performance for visual tasks.

Contribution

It proposes a novel global enhancement module and a deep reinforcement learning scheme to improve local feature extraction efficiency and robustness.

Findings

Achieves robustness against visual interference

Operates in real time

Outperforms existing methods on benchmarks

Abstract

Local feature provides compact and invariant image representation for various visual tasks. Current deep learning-based local feature algorithms always utilize convolution neural network (CNN) architecture with limited receptive field. Besides, even with high-performance GPU devices, the computational efficiency of local features cannot be satisfactory. In this paper, we tackle such problems by proposing a CNN-based local feature algorithm. The proposed method introduces a global enhancement module to fuse global visual clues in a light-weight network, and then optimizes the network by novel deep reinforcement learning scheme from the perspective of local feature matching task. Experiments on the public benchmarks demonstrate that the proposal can achieve considerable robustness against visual interference and meanwhile run in real time.