# Why-So-Deep: Towards Boosting Previously Trained Models for Visual Place Recognition

arXiv: https://arxiv.org/abs/2201.03212 · HTML: https://mlsanity.com/paper/2201.03212/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 23 of its 34 references that are in the map.
- First posted on arXiv 10 Jan 2022.

## Builds on (most cited first, 23 of 23)

- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [ORB-SLAM: a Versatile and Accurate Monocular SLAM System](https://mlsanity.com/paper/1502.00956/) (2015)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [From Coarse to Fine: Robust Hierarchical Localization at Large Scale](https://mlsanity.com/paper/1812.03506/) (2018)
- [VINS-Mono: A Robust and Versatile Monocular Visual-Inertial State Estimator](https://mlsanity.com/paper/1708.03852/) (2017)
- [Deep Learning Features at Scale for Visual Place Recognition](https://mlsanity.com/paper/1701.05105/) (2017)
- [The Application of Two-level Attention Models in Deep Convolutional Neural Network for Fine-grained Image Classification](https://mlsanity.com/paper/1411.6447/) (2014)
- [DGC-Net: Dense Geometric Correspondence Network](https://mlsanity.com/paper/1810.08393/) (2018)
- [X-View: Graph-Based Semantic Multi-View Localization](https://mlsanity.com/paper/1709.09905/) (2017)
- [maplab: An Open Framework for Research in Visual-inertial Mapping and Localization](https://mlsanity.com/paper/1711.10250/) (2017)
- [Attention-based Pyramid Aggregation Network for Visual Place Recognition](https://mlsanity.com/paper/1808.00288/) (2018)
- [Localizing Discriminative Visual Landmarks for Place Recognition](https://mlsanity.com/paper/1904.06635/) (2019)
- [STA-VPR: Spatio-temporal Alignment for Visual Place Recognition](https://mlsanity.com/paper/2103.13580/) (2021)
- [A Hierarchical Dual Model of Environment- and Place-Specific Utility for Visual Place Recognition](https://mlsanity.com/paper/2107.02440/) (2021)
- [DEDUCE: Diverse scEne Detection methods in Unseen Challenging Environments](https://mlsanity.com/paper/1908.00191/) (2019)
- [Multiple Hypothesis Semantic Mapping for Robust Data Association](https://mlsanity.com/paper/2012.04423/) (2020)
- [Spatio-Semantic ConvNet-Based Visual Place Recognition](https://mlsanity.com/paper/1909.07671/) (2019)
- [Loop-box: Multi-Agent Direct SLAM Triggered by Single Loop Closure for Large-Scale Mapping](https://mlsanity.com/paper/2009.13851/) (2020)
- [Geometric Image Correspondence Verification by Dense Pixel Matching](https://mlsanity.com/paper/1904.06882/) (2019)
- [Visual place recognition using landmark distribution descriptors](https://mlsanity.com/paper/1608.04274/) (2016)
- [Self-Supervised Visual Place Recognition Learning in Mobile Robots](https://mlsanity.com/paper/1905.04453/) (2019)
- [PCR-Pro: 3D Sparse and Different Scale Point Clouds Registration and Robust Estimation of Information Matrix For Pose Graph SLAM](https://mlsanity.com/paper/1808.09693/) (2018)
- [Learning Quintuplet Loss for Large-scale Visual Geo-Localization](https://mlsanity.com/paper/1907.11350/) (2019)

## Built on by (most cited first, 1 of 1)

- [GeoAdapt: Self-Supervised Test-Time Adaptation in LiDAR Place Recognition Using Geometric Priors](https://mlsanity.com/paper/2308.04638/) (2023)

## Abstract (arXiv)

Deep learning-based image retrieval techniques for the loop closure detection demonstrate satisfactory performance. However, it is still challenging to achieve high-level performance based on previously trained models in different geographical regions. This paper addresses the problem of their deployment with simultaneous localization and mapping (SLAM) systems in the new environment. The general baseline approach uses additional information, such as GPS, sequential keyframes tracking, and re-training the whole environment to enhance the recall rate. We propose a novel approach for improving image retrieval based on previously trained models. We present an intelligent method, MAQBOOL, to amplify the power of pre-trained models for better image recall and its application to real-time multiagent SLAM systems. We achieve comparable image retrieval results at a low descriptor dimension (512-D), compared to the high descriptor dimension (4096-D) of state-of-the-art methods. We use spatial information to improve the recall rate in image retrieval on pre-trained models.