# Tightly Coupled Learning Strategy for Weakly Supervised Hierarchical Place Recognition

arXiv: https://arxiv.org/abs/2202.06470 · HTML: https://mlsanity.com/paper/2202.06470/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 39 references that are in the map.
- First posted on arXiv 14 Feb 2022.

## Builds on (most cited first, 15 of 15)

- [Training data-efficient image transformers & distillation through attention](https://mlsanity.com/paper/2012.12877/) (2020)
- [SuperGlue: Learning Feature Matching with Graph Neural Networks](https://mlsanity.com/paper/1911.11763/) (2019)
- [SuperPoint: Self-Supervised Interest Point Detection and Description](https://mlsanity.com/paper/1712.07629/) (2017)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [ORB-SLAM2: an Open-Source SLAM System for Monocular, Stereo and RGB-D Cameras](https://mlsanity.com/paper/1610.06475/) (2016)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [D2-Net: A Trainable CNN for Joint Detection and Description of Local Features](https://mlsanity.com/paper/1905.03561/) (2019)
- [Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition](https://mlsanity.com/paper/2103.01486/) (2021)
- [On the Performance of ConvNet Features for Place Recognition](https://mlsanity.com/paper/1501.04158/) (2015)
- [SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition](https://mlsanity.com/paper/2102.11603/) (2021)
- [Detect-to-Retrieve: Efficient Regional Aggregation for Image Search](https://mlsanity.com/paper/1812.01584/) (2018)
- [shapeDTW: shape Dynamic Time Warping](https://mlsanity.com/paper/1606.01601/) (2016)
- [Single-View Place Recognition under Seasonal Changes](https://mlsanity.com/paper/1808.06516/) (2018)
- [Generalized Contrastive Optimization of Siamese Networks for Place Recognition](https://mlsanity.com/paper/2103.06638/) (2021)
- [STA-VPR: Spatio-temporal Alignment for Visual Place Recognition](https://mlsanity.com/paper/2103.13580/) (2021)

## Built on by (most cited first, 5 of 5)

- [Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition](https://mlsanity.com/paper/2402.14505/) (2024)
- [SelaVPR++: Towards Seamless Adaptation of Foundation Models for Efficient Place Recognition](https://mlsanity.com/paper/2502.16601/) (2025)
- [StructVPR: Distill Structural Knowledge with Weighting Samples for Visual Place Recognition](https://mlsanity.com/paper/2212.00937/) (2022)
- [AANet: Aggregation and Alignment Network with Semi-hard Positive Sample Mining for Hierarchical Place Recognition](https://mlsanity.com/paper/2310.05184/) (2023)
- [StructVPR++: Distill Structural and Semantic Knowledge with Weighting Samples for Visual Place Recognition](https://mlsanity.com/paper/2503.06601/) (2025)

## Abstract (arXiv)

Visual place recognition (VPR) is a key issue for robotics and autonomous systems. For the trade-off between time and performance, most of methods use the coarse-to-fine hierarchical architecture, which consists of retrieving top-N candidates using global features, and re-ranking top-N with local features. However, since the two types of features are usually processed independently, re-ranking may harm global retrieval, termed re-ranking confusion. Moreover, re-ranking is limited by global retrieval. In this paper, we propose a tightly coupled learning (TCL) strategy to train triplet models. Different from original triplet learning (OTL) strategy, it combines global and local descriptors for joint optimization. In addition, a bidirectional search dynamic time warping (BS-DTW) algorithm is also proposed to mine locally spatial information tailored to VPR in re-ranking. The experimental results on public benchmarks show that the models using TCL outperform the models using OTL, and TCL can be used as a general strategy to improve performance for weakly supervised ranking tasks. Further, our lightweight unified model is better than several state-of-the-art methods and has over an order of magnitude of computational efficiency to meet the real-time requirements of robots.