SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition

TL;DR

SeqNet introduces a hybrid approach for visual place recognition that combines short learned sequential descriptors with single-image descriptors, significantly improving initial match hypotheses and overall accuracy in challenging environments.

Contribution

The paper proposes a novel hybrid system using SeqNet for short sequence encoding and selective score aggregation, advancing the state-of-the-art in sequence-based place recognition.

Findings

Outperforms recent state-of-the-art methods on benchmark datasets.

Uses short learned sequential descriptors for high-quality initial hypotheses.

Demonstrates robustness in challenging environments.

Abstract

Visual Place Recognition (VPR) is the task of matching current visual imagery from a camera to images stored in a reference map of the environment. While initial VPR systems used simple direct image methods or hand-crafted visual features, recent work has focused on learning more powerful visual features and further improving performance through either some form of sequential matcher / filter or a hierarchical matching process. In both cases the performance of the initial single-image based system is still far from perfect, putting significant pressure on the sequence matching or (in the case of hierarchical systems) pose refinement stages. In this paper we present a novel hybrid system that creates a high performance initial match hypothesis generator using short learnt sequential descriptors, which enable selective control sequential score aggregation using single image learnt descriptors. Sequential descriptors are generated using a temporal convolutional network dubbed SeqNet, encoding short image sequences using 1-D convolutions, which are then matched against the corresponding temporal descriptors from the reference dataset to provide an ordered list of place match hypotheses. We then perform selective sequential score aggregation using shortlisted single image learnt descriptors from a separate pipeline to produce an overall place match hypothesis. Comprehensive experiments on challenging benchmark datasets demonstrate the proposed method outperforming recent state-of-the-art methods using the same amount of sequential information. Source code and supplementary material can be found at https://github.com/oravus/seqNet.