MultiRes-NetVLAD: Augmenting Place Recognition Training with Low-Resolution Imagery

02/18/2022
by   Ahmad Khaliq, et al.
6

Visual Place Recognition (VPR) is a crucial component of 6-DoF localization, visual SLAM and structure-from-motion pipelines, tasked to generate an initial list of place match hypotheses by matching global place descriptors. However, commonly-used CNN-based methods either process multiple image resolutions after training or use a single resolution and limit multi-scale feature extraction to the last convolutional layer during training. In this paper, we augment NetVLAD representation learning with low-resolution image pyramid encoding which leads to richer place representations. The resultant multi-resolution feature pyramid can be conveniently aggregated through VLAD into a single compact representation, avoiding the need for concatenation or summation of multiple patches in recent multi-scale approaches. Furthermore, we show that the underlying learnt feature tensor can be combined with existing multi-scale approaches to improve their baseline performance. Evaluation on 15 viewpoint-varying and viewpoint-consistent benchmarking datasets confirm that the proposed MultiRes-NetVLAD leads to state-of-the-art Recall@N performance for global descriptor based retrieval, compared against 11 existing techniques. Source code is publicly available at https://github.com/Ahmedest61/MultiRes-NetVLAD.

READ FULL TEXT

page 1

page 7

research
11/19/2019

HighEr-Resolution Network for Image Demosaicing and Enhancing

Neural-networks based image restoration methods tend to use low-resoluti...
research
09/20/2020

Scale-Localized Abstract Reasoning

We consider the abstract relational reasoning task, which is commonly us...
research
02/23/2021

SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition

Visual Place Recognition (VPR) is the task of matching current visual im...
research
01/01/2020

Multi-organ Segmentation over Partially Labeled Datasets with Multi-scale Feature Abstraction

This paper presents a unified training strategy that enables a novel mul...
research
02/29/2020

Augmenting Visual Place Recognition with Structural Cues

In this paper, we propose to augment image-based place recognition with ...
research
05/23/2018

SNIPER: Efficient Multi-Scale Training

We present SNIPER, an algorithm for performing efficient multi-scale tra...
research
02/16/2021

Learning deep multiresolution representations for pansharpening

Retaining spatial characteristics of panchromatic image and spectral inf...

Please sign up or login with your details

Forgot password? Click here to reset