MapFM: Foundation Model-Driven HD Mapping with Multi-Task Contextual Learning
This addresses the need for accurate HD maps in autonomous driving for localization, planning, and decision-making, though it appears incremental as it builds on existing foundation model and multi-task learning approaches.
The paper tackles the problem of online vectorized HD map generation for autonomous driving by introducing MapFM, an enhanced End-to-End model that incorporates a foundation model for camera image encoding and uses multi-task learning with auxiliary semantic segmentation heads. The approach results in significantly boosted feature representation quality and higher accuracy/improved quality of predicted vectorized HD maps.
In autonomous driving, high-definition (HD) maps and semantic maps in bird's-eye view (BEV) are essential for accurate localization, planning, and decision-making. This paper introduces an enhanced End-to-End model named MapFM for online vectorized HD map generation. We show significantly boost feature representation quality by incorporating powerful foundation model for encoding camera images. To further enrich the model's understanding of the environment and improve prediction quality, we integrate auxiliary prediction heads for semantic segmentation in the BEV representation. This multi-task learning approach provides richer contextual supervision, leading to a more comprehensive scene representation and ultimately resulting in higher accuracy and improved quality of the predicted vectorized HD maps. The source code is available at https://github.com/LIvanoff/MapFM.