MachMap: End-to-End Vectorized Solution for Compact HD-Map Construction (2306.10301v1)
Abstract: This report introduces the 1st place winning solution for the Autonomous Driving Challenge 2023 - Online HD-map Construction. By delving into the vectorization pipeline, we elaborate an effective architecture, termed as MachMap, which formulates the task of HD-map construction as the point detection paradigm in the bird-eye-view space with an end-to-end manner. Firstly, we introduce a novel map-compaction scheme into our framework, leading to reducing the number of vectorized points by 93% without any expression performance degradation. Build upon the above process, we then follow the general query-based paradigm and propose a strong baseline with integrating a powerful CNN-based backbone like InternImage, a temporal-based instance decoder and a well-designed point-mask coupling head. Additionally, an extra optional ensemble stage is utilized to refine model predictions for better performance. Our MachMap-tiny with IN-1K initialization achieves a mAP of 79.1 on the Argoverse2 benchmark and the further improved MachMap-huge reaches the best mAP of 83.5, outperforming all the other online HD-map construction approaches on the final leaderboard with a distinct performance margin (> 9.8 mAP at least).
- Deformable convolutional networks. In Proceedings of the IEEE international conference on computer vision, pages 764–773, 2017.
- Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition, pages 248–255. Ieee, 2009.
- Exploring recurrent long-term temporal fusion for multi-view 3d perception. arXiv, 2023.
- Bevdet: High-performance multi-camera 3d object detection in bird-eye-view. arXiv preprint arXiv:2112.11790, 2021.
- Dn-detr: Accelerate detr training by introducing query denoising. In CVPR, 2022.
- Mask dino: Towards a unified transformer-based framework for object detection and segmentation. In CVPR, 2023.
- Maptr: Structured modeling and learning for online vectorized hd map construction. arXiv preprint arXiv:2208.14437, 2022.
- Decoupled weight decay regularization. 2019.
- V-net: Fully convolutional neural networks for volumetric medical image segmentation. IEEE, 2016.
- End-to-end vectorized hd-map construction with piecewise bezier curve. In CVPR, 2023.
- The douglas-peucker algorithm for line simplification: re-evaluation through visualization. In Computer Graphics Forum, volume 9, pages 213–225. Wiley Online Library, 1990.
- Simplification and generalization of large scale data for roads: a comparison of two filtering algorithms. Cartography and Geographic Information Systems, 22(4):264–275, 1995.
- Internimage: Exploring large-scale vision foundation models with deformable convolutions. In CVPR, 2023.
- Argoverse 2: Next generation datasets for self-driving perception and forecasting. In Neural Information Processing Systems, 2021.
- Scene parsing through ade20k dataset. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 633–641, 2017.
- Bidirectional feature pyramid network with recurrent attention residual modules for shadow detection. In ECCV, 2018.