MapTR: Structured Modeling and Learning for Online Vectorized HD Map Construction

By Bencheng Liao, Shaoyu Chen, Xinggang Wang, Tianheng Cheng, Qian Zhang, Wenyu Liu, Chang Huang. International Conference on Learning Representations, 2022

416 citations recorded by the source corpus.

High-definition (HD) map provides abundant and precise environmental information of the driving scene, serving as a fundamental and indispensable component for planning in autonomous driving system. We present MapTR, a structured end-to-end Transformer for efficient online vectorized HD map construction. We propose a unified permutation-equivalent modeling approach, i.e., modeling map element as a point set with a group of equivalent permutations, which accurately describes the shape of map element and stabilizes the learning process. We design a hierarchical query embedding scheme to flexibly encode structured map information and perform hierarchical bipartite matching for map element learning. MapTR achieves the best performance and efficiency with only camera input among existing vectorized map construction approaches on nuScenes dataset. In particular, MapTR-nano runs at real-time inference speed ($25.1$ FPS) on RTX 3090, $8\times$ faster than the existing state-of-the-art camera-based method while achieving $5.0$ higher mAP. Even compared with the existing state-of-the-art multi-modality method, MapTR-nano achieves $0.7$ higher mAP, and MapTR-tiny achieves $13.5$ higher mAP and $3\times$ faster inference speed. Abundant qualitative results show that MapTR maintains stable and robust map construction quality in complex and various driving scenes. MapTR is of great application value in autonomous driving. Code and more demos are available at \url{https://github.com/hustvl/MapTR}.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/e26ea26f3195cfc7675f5ae3795b4960b76b97d4

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

1. Zhaoxiang Zhang
2. Qian Zhang
3. Bencheng Liao
4. Lue Fan
5. Wenyu Liu
6. Shaoyu Chen
7. Xinggang Wang
8. Yuntao Chen

In Computer Vision and Pattern Recognition

1. Xinggang Wang
2. Wenyu Liu
3. Qian Zhang
4. Zhaoxiang Zhang
5. Chang Huang
6. Helong Zhou
7. Bencheng Liao
8. Lue Fan

In Signal Processing

1. Zhaoxiang Zhang
2. Qian Zhang
3. Wenyu Liu
4. Xinggang Wang
5. Yuntao Chen
6. Chang Huang

Citations in the corpus

• VAD: Vectorized Scene Representation for Efficient Autonomous Driving (2023, background): “3.4).”
• Driving Into the Future: Multiview Visual Forecasting and Planning with World Model for Autonomous Driving (2023, methodology): “Additionally, we evaluate 3D object detection [62] and online map construction [38].”
• DiffusionDrive: Truncated Diffusion Model for End-to-End Autonomous Driving (2024, neutral): “End-to-end autonomous driving has gained significant attention in recent years due to advancements in perception models (detection [5, 17, 24, 42], tracking [54–56], online mapping [ 27, 28, 32], etc .)”
• VADv2: End-to-End Vectorized Autonomous Driving via Probabilistic Planning (2024, background): “MapTR [29, 30] introduces permutation equivalence and hierarchical matching strategies, significantly improving mapping performance.”
• Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability (2024, neutral): “Given that the detectors [82, 84] take image sequences as inputs, Drive-WM has to accomplish all steps of the denoising process before perception.”
• End-to-End Autonomous Driving: Challenges and Frontiers (2023, neutral): “To perform well on downstream tasks like planning, the objective to be optimized for the model needs to be sophisticated enough, beyond frame-level perception.”