DiffSemanticFusion: Semantic Raster BEV Fusion for Autonomous Driving via Online HD Map Diffusion

• URL: http://arxiv.org/abs/2508.01778v1
• Date: Sun, 03 Aug 2025 14:32:05 GMT
• Title: DiffSemanticFusion: Semantic Raster BEV Fusion for Autonomous Driving via Online HD Map Diffusion
• Authors: Zhigang Sun, Yiru Wang, Anqing Jiang, Shuo Wang, Yu Gao, Yuwen Heng, Shouyi Zhang, An He, Hao Jiang, Jinhao Chai, Zichong Gu, Wang Jijun, Shichen Tang, Lavdim Halilaj, Juergen Luettin, Hao Sun,
• Abstract summary: We propose DiffSemanticFusion -- a fusion framework for trajectory prediction and planning.<n>Our approach reasons over a semantic-fused BEV space, enhanced by a map diffusion module.<n>Experiments on real-world autonomous driving benchmarks, nuScenes and NAVSIM, demonstrate improved performance over several state-of-the-art methods.
• Score: 14.872416661028144
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Autonomous driving requires accurate scene understanding, including road geometry, traffic agents, and their semantic relationships. In online HD map generation scenarios, raster-based representations are well-suited to vision models but lack geometric precision, while graph-based representations retain structural detail but become unstable without precise maps. To harness the complementary strengths of both, we propose DiffSemanticFusion -- a fusion framework for multimodal trajectory prediction and planning. Our approach reasons over a semantic raster-fused BEV space, enhanced by a map diffusion module that improves both the stability and expressiveness of online HD map representations. We validate our framework on two downstream tasks: trajectory prediction and planning-oriented end-to-end autonomous driving. Experiments on real-world autonomous driving benchmarks, nuScenes and NAVSIM, demonstrate improved performance over several state-of-the-art methods. For the prediction task on nuScenes, we integrate DiffSemanticFusion with the online HD map informed QCNet, achieving a 5.1\% performance improvement. For end-to-end autonomous driving in NAVSIM, DiffSemanticFusion achieves state-of-the-art results, with a 15\% performance gain in NavHard scenarios. In addition, extensive ablation and sensitivity studies show that our map diffusion module can be seamlessly integrated into other vector-based approaches to enhance performance. All artifacts are available at https://github.com/SunZhigang7/DiffSemanticFusion.

Related papers

• RTMap: Real-Time Recursive Mapping with Change Detection and Localization [8.343318095882232]
  RTMap persistently crowdsourcing a multi-traversal HD map as a self-evolutional memory.<n>On onboard agents, RTMap simultaneously addresses three core challenges in an end-to-end fashion.<n> Experiments on several public autonomous driving datasets demonstrate our solid performance on both the prior-aided map quality and the localization accuracy.
  arXiv  Detail & Related papers  (2025-07-01T17:32:30Z)
• MapFM: Foundation Model-Driven HD Mapping with Multi-Task Contextual Learning [0.0]
  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.<n>This paper introduces an enhanced End-to-End model named MapFM for online vectorized HD map generation.
  arXiv  Detail & Related papers  (2025-06-18T09:42:30Z)
• Neural Semantic Map-Learning for Autonomous Vehicles [85.8425492858912]
  We present a mapping system that fuses local submaps gathered from a fleet of vehicles at a central instance to produce a coherent map of the road environment. Our method jointly aligns and merges the noisy and incomplete local submaps using a scene-specific Neural Signed Distance Field. We leverage memory-efficient sparse feature-grids to scale to large areas and introduce a confidence score to model uncertainty in scene reconstruction.
  arXiv  Detail & Related papers  (2024-10-10T10:10:03Z)
• DiFSD: Ego-Centric Fully Sparse Paradigm with Uncertainty Denoising and Iterative Refinement for Efficient End-to-End Self-Driving [55.53171248839489]
  We propose an ego-centric fully sparse paradigm, named DiFSD, for end-to-end self-driving.<n>Specifically, DiFSD mainly consists of sparse perception, hierarchical interaction and iterative motion planner.<n>Experiments conducted on nuScenes and Bench2Drive datasets demonstrate the superior planning performance and great efficiency of DiFSD.
  arXiv  Detail & Related papers  (2024-09-15T15:55:24Z)
• GenMapping: Unleashing the Potential of Inverse Perspective Mapping for Robust Online HD Map Construction [20.1127163541618]
  We have designed a universal map generation framework, GenMapping. The framework is established with a triadic synergy architecture, including principal and dual auxiliary branches. A thorough array of experimental results shows that the proposed model surpasses current state-of-the-art methods in both semantic mapping and vectorized mapping, while also maintaining a rapid inference speed.
  arXiv  Detail & Related papers  (2024-09-13T10:15:28Z)
• MFTraj: Map-Free, Behavior-Driven Trajectory Prediction for Autonomous Driving [15.965681867350215]
  This paper introduces a trajectory prediction model tailored for autonomous driving. It harnesses historical trajectory data combined with a novel geometric dynamic graph-based behavior-aware module. It achieves computational efficiency and reduced parameter overhead.
  arXiv  Detail & Related papers  (2024-05-02T13:13:52Z)
• Video Killed the HD-Map: Predicting Multi-Agent Behavior Directly From Aerial Images [14.689298253430568]
  We propose an aerial image-based map (AIM) representation that requires minimal annotation and provides rich road context information for traffic agents like pedestrians and vehicles. Our results demonstrate competitive multi-agent trajectory prediction performance especially for pedestrians in the scene when using our AIM representation.
  arXiv  Detail & Related papers  (2023-05-19T17:48:01Z)
• GoRela: Go Relative for Viewpoint-Invariant Motion Forecasting [121.42898228997538]
  We propose an efficient shared encoding for all agents and the map without sacrificing accuracy or generalization. We leverage pair-wise relative positional encodings to represent geometric relationships between the agents and the map elements in a heterogeneous spatial graph. Our decoder is also viewpoint agnostic, predicting agent goals on the lane graph to enable diverse and context-aware multimodal prediction.
  arXiv  Detail & Related papers  (2022-11-04T16:10:50Z)
• Multi-Modal Fusion Transformer for End-to-End Autonomous Driving [59.60483620730437]
  We propose TransFuser, a novel Multi-Modal Fusion Transformer, to integrate image and LiDAR representations using attention. Our approach achieves state-of-the-art driving performance while reducing collisions by 76% compared to geometry-based fusion.
  arXiv  Detail & Related papers  (2021-04-19T11:48:13Z)
• Learning Lane Graph Representations for Motion Forecasting [92.88572392790623]
  We construct a lane graph from raw map data to preserve the map structure. We exploit a fusion network consisting of four types of interactions, actor-to-lane, lane-to-lane, lane-to-actor and actor-to-actor. Our approach significantly outperforms the state-of-the-art on the large scale Argoverse motion forecasting benchmark.
  arXiv  Detail & Related papers  (2020-07-27T17:59:49Z)
• VectorNet: Encoding HD Maps and Agent Dynamics from Vectorized Representation [74.56282712099274]
  This paper introduces VectorNet, a hierarchical graph neural network that exploits the spatial locality of individual road components represented by vectors. By operating on the vectorized high definition (HD) maps and agent trajectories, we avoid lossy rendering and computationally intensive ConvNet encoding steps. We evaluate VectorNet on our in-house behavior prediction benchmark and the recently released Argoverse forecasting dataset.
  arXiv  Detail & Related papers  (2020-05-08T19:07:03Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.