# MapNeXt: Revisiting Training and Scaling Practices for Online Vectorized HD Map Construction

arXiv: https://arxiv.org/abs/2401.07323 · HTML: https://mlsanity.com/paper/2401.07323/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 46 of its 55 references that are in the map.
- First posted on arXiv 14 Jan 2024.

## Builds on (most cited first, 30 of 46)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [End-to-End Object Detection with Transformers](https://mlsanity.com/paper/2005.12872/) (2020)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [Feature Pyramid Networks for Object Detection](https://mlsanity.com/paper/1612.03144/) (2016)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Deformable DETR: Deformable Transformers for End-to-End Object Detection](https://mlsanity.com/paper/2010.04159/) (2020)
- [BEiT: BERT Pre-Training of Image Transformers](https://mlsanity.com/paper/2106.08254/) (2021)
- [Masked-attention Mask Transformer for Universal Image Segmentation](https://mlsanity.com/paper/2112.01527/) (2021)
- [PointPillars: Fast Encoders for Object Detection from Point Clouds](https://mlsanity.com/paper/1812.05784/) (2018)
- [Cascade R-CNN: Delving into High Quality Object Detection](https://mlsanity.com/paper/1712.00726/) (2017)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [BEVFormer: Learning Bird's-Eye-View Representation from Multi-Camera Images via Spatiotemporal Transformers](https://mlsanity.com/paper/2203.17270/) (2022)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Center-based 3D Object Detection and Tracking](https://mlsanity.com/paper/2006.11275/) (2020)
- [Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3D](https://mlsanity.com/paper/2008.05711/) (2020)
- [Scaling Vision Transformers](https://mlsanity.com/paper/2106.04560/) (2021)
- [Deep Networks with Stochastic Depth](https://mlsanity.com/paper/1603.09382/) (2016)
- [COCO-Stuff: Thing and Stuff Classes in Context](https://mlsanity.com/paper/1612.03716/) (2016)
- [BEVDet: High-performance Multi-camera 3D Object Detection in Bird-Eye-View](https://mlsanity.com/paper/2112.11790/) (2021)
- [Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecasting](https://mlsanity.com/paper/2301.00493/) (2023)
- [DETR3D: 3D Object Detection from Multi-view Images via 3D-to-2D Queries](https://mlsanity.com/paper/2110.06922/) (2021)
- [How to train your ViT? Data, Augmentation, and Regularization in Vision Transformers](https://mlsanity.com/paper/2106.10270/) (2021)
- [PETR: Position Embedding Transformation for Multi-View 3D Object Detection](https://mlsanity.com/paper/2203.05625/) (2022)

## Built on by (most cited first, 6 of 6)

- [Is Your HD Map Constructor Reliable under Sensor Corruptions?](https://mlsanity.com/paper/2406.12214/) (2024)
- [MapExpert: Online HD Map Construction with Simple and Efficient Sparse Map Element Expert](https://mlsanity.com/paper/2412.12704/) (2024)
- [SafeMap: Robust HD Map Construction from Incomplete Observations](https://mlsanity.com/paper/2507.00861/) (2025)
- [PrevPredMap: Exploring Temporal Modeling with Previous Predictions for Online Vectorized HD Map Construction](https://mlsanity.com/paper/2407.17378/) (2024)
- [SparseMeXT Unlocking the Potential of Sparse Representations for HD Map Construction](https://mlsanity.com/paper/2505.08808/) (2025)
- [FastMap: Fast Queries Initialization Based Vectorized HD Map Reconstruction Framework](https://mlsanity.com/paper/2503.05492/) (2025)

## Abstract (arXiv)

High-Definition (HD) maps are pivotal to autopilot navigation. Integrating the capability of lightweight HD map construction at runtime into a self-driving system recently emerges as a promising direction. In this surge, vision-only perception stands out, as a camera rig can still perceive the stereo information, let alone its appealing signature of portability and economy. The latest MapTR architecture solves the online HD map construction task in an end-to-end fashion but its potential is yet to be explored. In this work, we present a full-scale upgrade of MapTR and propose MapNeXt, the next generation of HD map learning architecture, delivering major contributions from the model training and scaling perspectives. After shedding light on the training dynamics of MapTR and exploiting the supervision from map elements thoroughly, MapNeXt-Tiny raises the mAP of MapTR-Tiny from 49.0% to 54.8%, without any architectural modifications. Enjoying the fruit of map segmentation pre-training, MapNeXt-Base further lifts the mAP up to 63.9% that has already outperformed the prior art, a multi-modality MapTR, by 1.4% while being ~1.8× faster. Towards pushing the performance frontier to the next level, we draw two conclusions on practical model scaling: increased query favors a larger decoder network for adequate digestion; a large backbone steadily promotes the final accuracy without bells and whistles. Building upon these two rules of thumb, MapNeXt-Huge achieves state-of-the-art performance on the challenging nuScenes benchmark. Specifically, we push the mapless vision-only single-model performance to be over 78% for the first time, exceeding the best model from existing methods by 16%.