# TRAIL: Trajectory-Aware Visual Place Recognition against Unordered Databases

arXiv: https://arxiv.org/abs/2609.07373 · HTML: https://mlsanity.com/paper/2609.07373/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 47 references that are in the map.
- First posted on arXiv 7 Sep 2026.

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

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [NetVLAD: CNN architecture for weakly supervised place recognition](https://mlsanity.com/paper/1511.07247/) (2015)
- [Fine-tuning CNN Image Retrieval with No Human Annotation](https://mlsanity.com/paper/1711.02512/) (2017)
- [AnyLoc: Towards Universal Visual Place Recognition](https://mlsanity.com/paper/2308.00688/) (2023)
- [MixVPR: Feature Mixing for Visual Place Recognition](https://mlsanity.com/paper/2303.02190/) (2023)
- [Optimal Transport Aggregation for Visual Place Recognition](https://mlsanity.com/paper/2311.15937/) (2023)
- [BoQ: A Place is Worth a Bag of Learnable Queries](https://mlsanity.com/paper/2405.07364/) (2024)
- [MegaLoc: One Retrieval to Place Them All](https://mlsanity.com/paper/2502.17237/) (2025)
- [SeqNet: Learning Descriptors for Sequence-based Hierarchical Place Recognition](https://mlsanity.com/paper/2102.11603/) (2021)
- [4Seasons: A Cross-Season Dataset for Multi-Weather SLAM in Autonomous Driving](https://mlsanity.com/paper/2009.06364/) (2020)
- [Towards Seamless Adaptation of Pre-trained Models for Visual Place Recognition](https://mlsanity.com/paper/2402.14505/) (2024)
- [Close, But Not There: Boosting Geographic Distance Sensitivity in Visual Place Recognition](https://mlsanity.com/paper/2407.02422/) (2024)
- [Learning Sequential Descriptors for Sequence-based Visual Place Recognition](https://mlsanity.com/paper/2207.03868/) (2022)
- [Focus on Local: Finding Reliable Discriminative Regions for Visual Place Recognition](https://mlsanity.com/paper/2504.09881/) (2025)
- [Pair-VPR: Place-Aware Pre-training and Contrastive Pair Classification for Visual Place Recognition with Vision Transformers](https://mlsanity.com/paper/2410.06614/) (2024)
- [JIST: Joint Image and Sequence Training for Sequential Visual Place Recognition](https://mlsanity.com/paper/2403.19787/) (2024)
- [Probabilistic Appearance-Invariant Topometric Localization with New Place Awareness](https://mlsanity.com/paper/2107.07707/) (2021)
- [Learning Sequence Descriptor based on Spatio-Temporal Attention for Visual Place Recognition](https://mlsanity.com/paper/2305.11467/) (2023)
- [4Seasons: Benchmarking Visual SLAM and Long-Term Localization for Autonomous Driving in Challenging Conditions](https://mlsanity.com/paper/2301.01147/) (2022)
- [ROVER: Robust Loop Closure Verification with Trajectory Prior in Repetitive Environments](https://mlsanity.com/paper/2508.13488/) (2025)

## Abstract (arXiv)

Modern Visual Place Recognition (VPR) methods excel on standard benchmarks yet remain brittle in feature-poor environments. By treating each query image in isolation, they discard the sequential context in any real trajectory. We formalize a task that exploits this context: given a query sequence, localize the final image against an unordered reference database -- which, unlike sequence-to-sequence methods, requires no sequential structure in the database. We propose TRAIL (TRajectory-Aware Image Localization), a principled framework based on Conditional Random Fields (CRF) that combines learned functions for visual similarity and for camera-motion consistency, refining a distribution over candidate references as each query arrives. A lightweight post-processing layer atop any pre-trained VPR backbone, TRAIL improves a state-of-the-art baseline by up to 8.3 percentage points on our primary benchmark, transfers to unseen datasets without retraining, and delivers its largest gains where visual cues are scarce.