RefDiffMap: Diffusion-Guided Progressive Refinement for Vectorized HD Map Construction
- 12 January 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 11 (3) , 2554-2561
Abstract
High-definition (HD) map learning serves as an essential component of autonomous driving scene understanding, providing structured priors for planning and prediction. Recent transformer-based methods regress vectorized map elements via deformable attention over Bird’s-Eye View (BEV) features. They typically employ a single-pass paradigm, starting from a set of initial queries. However, these queries struggle to precisely localize map elements within the large-scale BEV space. This difficulty is severely amplified when using lightweight backbones that produce less distinctive features. To address this, we propose RefDiffMap, which recasts map construction as a progressive refinement process driven by a diffusion model. We introduce a novel denoising query generator that, at each step, leverages the intermediate noisy geometry to sample relevant features from adaptive BEV RoIs. These features are distilled into context-aware queries that guide the decoder’s next refinement. This creates a powerful geometry-feature co-evolution loop, allowing the model to iteratively correct localization errors. Comprehensive experiments show that RefDiffMap achieves competitive performance on the nuScenes and Argoverse 2 datasets. Notably, its robustness is highlighted with a ResNet-18 backbone, where it improves mAP by a significant 11.3% over our baseline MapTRv2. Further ablation studies validate the effectiveness of our approach.Keywords
Funding Information
- National Natural Science Foundation of China (62088102)
This publication has 26 references indexed in Scilit:
- Stream Query Denoising for Vectorized HD-Map Construction2024
- Online Vectorized HD Map Construction Using Geometry2024
- Diff3DETR: Agent-Based Diffusion Model for Semi-supervised 3D Object Detection2024
- MapTRv2: An End-to-End Framework for Online Vectorized HD Map ConstructionInternational Journal of Computer Vision, 2024
- MapTracker: Tracking with Strided Memory Fusion for Consistent Vector HD Mapping2024
- Diffusion-Based 3D Object Detection with Random Boxes2023
- DN-DETR: Accelerate DETR Training by Introducing Query DeNoising2022
- Accelerating DETR Convergence via Semantic-Aligned Matching2022
- nuScenes: A Multimodal Dataset for Autonomous Driving2020
- Deep Residual Learning for Image Recognition2016