# Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix It

arXiv: https://arxiv.org/abs/2312.06420 · HTML: https://mlsanity.com/paper/2312.06420/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 42 references that are in the map.
- First posted on arXiv 11 Dec 2023.

## Builds on (most cited first, 30 of 34)

- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [RandAugment: Practical automated data augmentation with a reduced search space](https://mlsanity.com/paper/1909.13719/) (2019)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers](https://mlsanity.com/paper/2203.17270/) (2022)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representation](https://mlsanity.com/paper/2205.13542/) (2022)
- [Argoverse: 3D Tracking and Forecasting with Rich Maps](https://mlsanity.com/paper/1911.02620/) (2019)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [ST-P3: End-to-end Vision-based Autonomous Driving via Spatial-Temporal Feature Learning](https://mlsanity.com/paper/2207.07601/) (2022)
- [MapTR: Structured Modeling and Learning for Online Vectorized HD Map Construction](https://mlsanity.com/paper/2208.14437/) (2022)
- [HDMapNet: An Online HD Map Construction and Evaluation Framework](https://mlsanity.com/paper/2107.06307/) (2021)
- [Cross-view Transformers for real-time Map-view Semantic Segmentation](https://mlsanity.com/paper/2205.02833/) (2022)
- [VectorMapNet: End-to-end Vectorized HD Map Learning](https://mlsanity.com/paper/2206.08920/) (2022)
- [MapTRv2: An End-to-End Framework for Online Vectorized HD Map Construction](https://mlsanity.com/paper/2308.05736/) (2023)
- [BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Driving](https://mlsanity.com/paper/2205.09743/) (2022)
- [Predicting Semantic Map Representations from Images using Pyramid Occupancy Networks](https://mlsanity.com/paper/2003.13402/) (2020)
- [M2BEV: Multi-Camera Joint 3D Detection and Segmentation with Unified Birds-Eye View Representation](https://mlsanity.com/paper/2204.05088/) (2022)
- [PolarFormer: Multi-camera 3D Object Detection with Polar Transformer](https://mlsanity.com/paper/2206.15398/) (2022)
- [StreamMapNet: Streaming Mapping Network for Vectorized Online HD Map Construction](https://mlsanity.com/paper/2308.12570/) (2023)
- [Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?](https://mlsanity.com/paper/2206.07959/) (2022)
- [BEVSegFormer: Bird's Eye View Semantic Segmentation From Arbitrary Camera Rigs](https://mlsanity.com/paper/2203.04050/) (2022)
- [Translating Images into Maps](https://mlsanity.com/paper/2110.00966/) (2021)
- [Structured Bird's-Eye-View Traffic Scene Understanding from Onboard Images](https://mlsanity.com/paper/2110.01997/) (2021)
- [Zenseact Open Dataset: A large-scale and diverse multimodal dataset for autonomous driving](https://mlsanity.com/paper/2305.02008/) (2023)
- [Efficient and Robust 2D-to-BEV Representation Learning via Geometry-guided Kernel Transformer](https://mlsanity.com/paper/2206.04584/) (2022)
- [InstaGraM: Instance-level Graph Modeling for Vectorized HD Map Learning](https://mlsanity.com/paper/2301.04470/) (2023)
- [SuperFusion: Multilevel LiDAR-Camera Fusion for Long-Range HD Map Generation](https://mlsanity.com/paper/2211.15656/) (2022)

## Built on by (most cited first, 30 of 44)

- [MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping](https://mlsanity.com/paper/2403.15951/) (2024)
- [OmniHD-Scenes: A Next-Generation Multimodal Dataset for Autonomous Driving](https://mlsanity.com/paper/2412.10734/) (2024)
- [PreSight: Enhancing Autonomous Vehicle Perception with City-Scale NeRF Priors](https://mlsanity.com/paper/2403.09079/) (2024)
- [Are NeRFs ready for autonomous driving? Towards closing the real-to-simulation gap](https://mlsanity.com/paper/2403.16092/) (2024)
- [SMART: Advancing Scalable Map Priors for Driving Topology Reasoning](https://mlsanity.com/paper/2502.04329/) (2025)
- [Exploring Real World Map Change Generalization of Prior-Informed HD Map Prediction Models](https://mlsanity.com/paper/2406.01961/) (2024)
- [HisTrackMap: Global Vectorized High-Definition Map Construction via History Map Tracking](https://mlsanity.com/paper/2503.07168/) (2025)
- [SDTagNet: Leveraging Text-Annotated Navigation Maps for Online HD Map Construction](https://mlsanity.com/paper/2506.08997/) (2025)
- [Unlocking Multimodal Document Intelligence: From Current Triumphs to Future Frontiers of Visual Document Retrieval](https://mlsanity.com/paper/2602.19961/) (2026)
- [Stability Under Scrutiny: Benchmarking Representation Paradigms for Online HD Mapping](https://mlsanity.com/paper/2510.10660/) (2025)
- [Exploring Semi-Supervised Learning for Online Mapping](https://mlsanity.com/paper/2410.10279/) (2024)
- [Contrasting local and global modeling with machine learning and satellite data: A case study estimating tree canopy height in African savannas](https://mlsanity.com/paper/2411.14354/) (2024)
- [Towards Foundational Models for Single-Chip Radar](https://mlsanity.com/paper/2509.12482/) (2025)
- [Enhancing Lane Segment Perception and Topology Reasoning with Crowdsourcing Trajectory Priors](https://mlsanity.com/paper/2411.17161/) (2024)
- [PseudoMapTrainer: Learning Online Mapping without HD Maps](https://mlsanity.com/paper/2508.18788/) (2025)
- [GASP: Unifying Geometric and Semantic Self-Supervised Pre-training for Autonomous Driving](https://mlsanity.com/paper/2503.15672/) (2025)
- [ExelMap: Explainable Element-based HD-Map Change Detection and Update](https://mlsanity.com/paper/2409.10178/) (2024)
- [M3TR: A Generalist Model for Real-World HD Map Completion](https://mlsanity.com/paper/2411.10316/) (2024)
- [MapGS: Generalizable Pretraining and Data Augmentation for Online Mapping via Novel View Synthesis](https://mlsanity.com/paper/2501.06660/) (2025)
- [Uncertainty-Instructed Structure Injection for Generalizable HD Map Construction](https://mlsanity.com/paper/2503.23109/) (2025)
- [Coherent Online Road Topology Estimation and Reasoning with Standard-Definition Maps](https://mlsanity.com/paper/2507.01397/) (2025)
- [Inferring Driving Maps by Deep Learning-based Trail Map Extraction](https://mlsanity.com/paper/2505.10258/) (2025)
- [MapRF: Weakly Supervised Online HD Map Construction via NeRF-Guided Self-Training](https://mlsanity.com/paper/2511.19527/) (2025)
- [nuScenes Revisited: Progress and Challenges in Autonomous Driving](https://mlsanity.com/paper/2512.02448/) (2025)
- [Bridging Structure and Language: Graph-Based Visual Reasoning for Autonomous Road Understanding](https://mlsanity.com/paper/2605.20942/) (2026)
- [D2HDMap: Non-visible Driveline Map Prior for Online Vectorized HD Map Prediction](https://mlsanity.com/paper/2606.20725/) (2026)
- [Epipolar Attention Field Transformers for Bird's Eye View Semantic Segmentation](https://mlsanity.com/paper/2412.01595/) (2024)
- [ArgoTweak: Towards Self-Updating HD Maps through Structured Priors](https://mlsanity.com/paper/2509.08764/) (2025)
- [TopoMaskV3: 3D Mask Head with Dense Offset and Height Predictions for Road Topology Understanding](https://mlsanity.com/paper/2603.01558/) (2026)
- [Impact of Localization Errors on Label Quality for Online HD Map Construction](https://mlsanity.com/paper/2603.03452/) (2026)

## Abstract (arXiv)

The task of online mapping is to predict a local map using current sensor observations, e.g. from lidar and camera, without relying on a pre-built map. State-of-the-art methods are based on supervised learning and are trained predominantly using two datasets: nuScenes and Argoverse 2. However, these datasets revisit the same geographic locations across training, validation, and test sets. Specifically, over 80% of nuScenes and 40% of Argoverse 2 validation and test samples are less than 5 m from a training sample. At test time, the methods are thus evaluated more on how well they localize within a memorized implicit map built from the training data than on extrapolating to unseen locations. Naturally, this data leakage causes inflated performance numbers and we propose geographically disjoint data splits to reveal the true performance in unseen environments. Experimental results show that methods perform considerably worse, some dropping more than 45 mAP, when trained and evaluated on proper data splits. Additionally, a reassessment of prior design choices reveals diverging conclusions from those based on the original split. Notably, the impact of lifting methods and the support from auxiliary tasks (e.g., depth supervision) on performance appears less substantial or follows a different trajectory than previously perceived. Splits can be found at https://github.com/LiljaAdam/geographical-splits