# Self-Supervised Place Recognition by Refining Temporal and Featural Pseudo Labels from Panoramic Data

arXiv: https://arxiv.org/abs/2208.09315 · HTML: https://mlsanity.com/paper/2208.09315/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 59 references that are in the map.
- First posted on arXiv 19 Aug 2022.

## Builds on (most cited first, 21 of 21)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [Representation Learning with Contrastive Predictive Coding](https://mlsanity.com/paper/1807.03748/) (2018)
- [Momentum Contrast for Unsupervised Visual Representation Learning](https://mlsanity.com/paper/1911.05722/) (2019)
- [Habitat: A Platform for Embodied AI Research](https://mlsanity.com/paper/1904.01201/) (2019)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning](https://mlsanity.com/paper/1704.03976/) (2017)
- [AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty](https://mlsanity.com/paper/1912.02781/) (2019)
- [Learning from Noisy Labels with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2007.08199/) (2020)
- [KITTI-360: A Novel Dataset and Benchmarks for Urban Scene Understanding in 2D and 3D](https://mlsanity.com/paper/2109.13410/) (2021)
- [From Coarse to Fine: Robust Hierarchical Localization at Large Scale](https://mlsanity.com/paper/1812.03506/) (2018)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [Gibson Env: Real-World Perception for Embodied Agents](https://mlsanity.com/paper/1808.10654/) (2018)
- [Confidence Regularized Self-Training](https://mlsanity.com/paper/1908.09822/) (2019)
- [Semi-parametric Topological Memory for Navigation](https://mlsanity.com/paper/1803.00653/) (2018)
- [PPFNet: Global Context Aware Local Features for Robust 3D Point Matching](https://mlsanity.com/paper/1802.02669/) (2018)
- [VPR-Bench: An Open-Source Visual Place Recognition Evaluation Framework with Quantifiable Viewpoint and Appearance Change](https://mlsanity.com/paper/2005.08135/) (2020)
- [Lightweight Unsupervised Deep Loop Closure](https://mlsanity.com/paper/1805.07703/) (2018)
- [Self-supervised Geometric Perception](https://mlsanity.com/paper/2103.03114/) (2021)
- [Simultaneous Edge Alignment and Learning](https://mlsanity.com/paper/1808.01992/) (2018)
- [NYU-VPR: Long-Term Visual Place Recognition Benchmark with View Direction and Data Anonymization Influences](https://mlsanity.com/paper/2110.09004/) (2021)
- [Self-Supervised Domain Calibration and Uncertainty Estimation for Place Recognition](https://mlsanity.com/paper/2203.04446/) (2022)

## Built on by (most cited first, 3 of 3)

- [DeepMapping2: Self-Supervised Large-Scale LiDAR Map Optimization](https://mlsanity.com/paper/2212.06331/) (2022)
- [Co-VisiON: Co-Visibility ReasONing on Sparse Image Sets of Indoor Scenes](https://mlsanity.com/paper/2506.16805/) (2025)
- [The Overlooked Value of Test-time Reference Sets in Visual Place Recognition](https://mlsanity.com/paper/2510.03751/) (2025)

## Abstract (arXiv)

Visual place recognition (VPR) using deep networks has achieved state-of-the-art performance. However, most of them require a training set with ground truth sensor poses to obtain positive and negative samples of each observation's spatial neighborhood for supervised learning. When such information is unavailable, temporal neighborhoods from a sequentially collected data stream could be exploited for self-supervised training, although we find its performance suboptimal. Inspired by noisy label learning, we propose a novel self-supervised framework named TF-VPR that uses temporal neighborhoods and learnable feature neighborhoods to discover unknown spatial neighborhoods. Our method follows an iterative training paradigm which alternates between: (1) representation learning with data augmentation, (2) positive set expansion to include the current feature space neighbors, and (3) positive set contraction via geometric verification. We conduct auto-labeling and generalization tests on both simulated and real datasets, with either RGB images or point clouds as inputs. The results show that our method outperforms self-supervised baselines in recall rate, robustness, and heading diversity, a novel metric we propose for VPR. Our code and datasets can be found at https://ai4ce.github.io/TF-VPR/