Addressing Diverging Training Costs Using BEVRestore for High-Resolution Bird's Eye View Map Construction
- 3 October 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (11) , 10700-10707
Abstract
Recent advancements in Bird's Eye View (BEV) fusion for map construction have demonstrated remarkable mapping of urban environments. However, their deep and bulky architecture incurs substantial amounts of backpropagation memory and computing latency. Consequently, the problem poses an unavoidable bottleneck in constructing high-resolution (HR) BEV maps, as their large-sized features cause significant increases in costs including GPU memory consumption and computing latency, named diverging training costs issue. Affected by the problem, most existing methods adopt low-resolution (LR) BEV and struggle to estimate the precise locations of urban scene components like road lanes, and sidewalks. As the imprecision leads to risky motion planning like collision avoidance, the diverging training costs issue has to be resolved. In this paper, we address the issue with our novel BEVRestore mechanism. Specifically, our proposed model encodes the features of each sensor to LR BEV space and restores them to HR space to establish a memory-efficient map constructor. To this end, we introduce the BEV restoration strategy, which restores aliasing, and blocky artifacts of the up-scaled BEV features, and narrows down the width of the labels. Our extensive experiments show that the proposed mechanism provides a plug-and-play, memory-efficient pipeline, enabling an HR map construction with a broad BEV scope. Our code will be publicly released.Keywords
All Related Versions
This publication has 28 references indexed in Scilit:
- BEVFormer v2: Adapting Modern Image Backbones to Bird's-Eye-View Recognition via Perspective Supervision2023
- End-to-End Vectorized HD-map Construction with Piecewise Bézier Curve2023
- BEV-Guided Multi-Modality Fusion for Driving Perception2023
- LiDAR2Map: In Defense of LiDAR-Based Semantic Map Construction Using Online Camera Distillation2023
- BEVFormer: Learning Bird’s-Eye-View Representation from Multi-camera Images via Spatiotemporal Transformers2022
- Cross-view Transformers for real-time Map-view Semantic Segmentation2022
- nuScenes: A Multimodal Dataset for Autonomous Driving2020
- PointPainting: Sequential Fusion for 3D Object Detection2020
- Feature Pyramid Networks for Object Detection2017
- Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network2016