DiffMap: Enhancing Map Segmentation With Map Prior Using Diffusion Model
- 6 September 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (11) , 9836-9843
Abstract
Constructing high-definition (HD) maps is a crucial requirement for enabling autonomous driving. In recent years, several map segmentation algorithms have been developed to address this need, leveraging advancements in Bird's-Eye View (BEV) perception. However, existing models still encounter challenges in producing realistic and consistent semantic map layouts. A prominent issue is the limited utilization of structured priors inherent in map segmentation masks. In light of this, we propose DiffMap, a novel approach specifically designed to model the structured priors of map segmentation masks using latent diffusion model. By incorporating this technique, the performance of existing semantic segmentation methods can be significantly enhanced and certain structural errors present in the segmentation outputs can be effectively rectified. Notably, the proposed module can be seamlessly integrated into any map segmentation model, thereby augmenting its capability to accurately delineate semantic information. Furthermore, through extensive visualization analysis, our model demonstrates superior proficiency in generating results that more accurately reflect real-world map layouts, further validating its efficacy in improving the quality of the generated maps.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (U22A20104, 52102464)
- Beijing Natural Science Foundation (L231008, L243008)
- Young Elite Scientist Sponsorship Program by Bast (BYESS2022153)
- State-funded postdoctoral researcher program of China (GZC20231285)
- Tsinghua University-Daimler Joint Research Center for Future Mobility and Tsinghua University-Didi Joint Research Center
This publication has 18 references indexed in Scilit:
- Adding Conditional Control to Text-to-Image Diffusion Models2023
- Neural Map Prior for Autonomous Driving2023
- MotionDiffuser: Controllable Multi-Agent Motion Prediction Using Diffusion2023
- High-Resolution Image Synthesis with Latent Diffusion Models2022
- Lift, Splat, Shoot: Encoding Images from Arbitrary Camera Rigs by Implicitly Unprojecting to 3D2020
- nuScenes: A Multimodal Dataset for Autonomous Driving2020
- PointPillars: Fast Encoders for Object Detection From Point Clouds2019
- Deep Residual Learning for Image Recognition2016
- U-Net: Convolutional Networks for Biomedical Image Segmentation2015
- Fully convolutional networks for semantic segmentation2015