mlsanityData to 14 Oct 2026

LIE: LiDAR-only HD Map Construction with Intensity Enhancement via Online Knowledge Distillation

arXiv 2605.01478 ↗First posted 2 May 2026cs.CV cs.AI cs.LGcited by 0

On the map: Knowledge distillation → Detection and dense prediction distillation

Abstract (arXiv)

Online High-Definition (HD) map construction is a key component of autonomous driving. Recent methods rely on multi-view camera images for cost-effective HD map segmentation, but cameras lack depth information for accurate scene geometry. In contrast, LiDAR provides precise 3D measurements but lacks dense semantic cues. In this work, we propose LIE, LiDAR-only semantic map construction method that employ Knowledge Distillation (KD) to handle the lack of dense semantic and texture cues. Specifically, the teacher branch fuses student LiDAR features and the corresponding 2D intensity map tile to provide dense supervision for segmenting map elements using online distillation scheme. Experimental results show that our method outperforms all single-modality approaches, achieving 8.2% higher mIoU than the state-of-the-art camera-based model on nuScenes. LIE is robust over long ranges and under challenging weather and lighting, and efficiently adapts to Argoverse2 with only 10% fine-tuning, surpassing camera-based models trained on the full dataset. Source code will be available hrefhttps://iv.ee.hm.edu/lie/here.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Detection and dense prediction distillation, one of 12 directions in Knowledge distillation. Builds on 41 of its 49 references that are in the map. First posted on arXiv 2 May 2026.

Builds on

41 of 49 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024MapDistill: Boosting Efficient Camera-based HD Map Construction via Camera-LiDAR Fusion Model Distillation16 Jul
  2. 2023DistillBEV: Boosting Multi-Camera 3D Object Detection with Cross-Modal Knowledge Distillation26 Sep
  3. 2023UniDistill: A Universal Cross-Modality Knowledge Distillation Framework for 3D Object Detection in Bird's-Eye View27 Mar
  4. 2022BEVDistill: Cross-Modal BEV Distillation for Multi-View 3D Object Detection17 Nov

Other references · 37

  1. 2014Adam: A Method for Stochastic Optimizationcited by 37,240
  2. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  3. 2015Distilling the Knowledge in a Neural Networkcited by 10,324
  4. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,248
  5. 2016Feature Pyramid Networks for Object Detectioncited by 5,036
  6. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  7. 2018PointPillars: Fast Encoders for Object Detection from Point Cloudscited by 1,715
  8. 2020Lift, Splat, Shoot: Encoding Images From Arbitrary Camera Rigs by Implicitly Unprojecting to 3Dcited by 1,044
  9. 2022BEVFusion: Multi-Task Multi-Sensor Fusion with Unified Bird's-Eye View Representationcited by 850
  10. 2023Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecastingcited by 707
  11. 2022PETR: Position Embedding Transformation for Multi-View 3D Object Detectioncited by 552
  12. 2022MapTR: Structured Modeling and Learning for Online Vectorized HD Map Constructioncited by 314
25 more
  1. 2021HDMapNet: An Online HD Map Construction and Evaluation Frameworkcited by 281
  2. 2022Cross-view Transformers for real-time Map-view Semantic Segmentationcited by 264
  3. 2022VectorMapNet: End-to-end Vectorized HD Map Learningcited by 245
  4. 2023MapTRv2: An End-to-End Framework for Online Vectorized HD Map Constructioncited by 193
  5. 2022BEVerse: Unified Perception and Prediction in Birds-Eye-View for Vision-Centric Autonomous Drivingcited by 191
  6. 2023StreamMapNet: Streaming Mapping Network for Vectorized Online HD Map Constructioncited by 145
  7. 2019xMUDA: Cross-Modal Unsupervised Domain Adaptation for 3D Semantic Segmentationcited by 129
  8. 2022Simple-BEV: What Really Matters for Multi-Sensor BEV Perception?cited by 128
  9. 2022BEVSegFormer: Bird's Eye View Semantic Segmentation From Arbitrary Camera Rigscited by 123
  10. 2020Attentional Feature Fusioncited by 115
  11. 2023Neural Map Prior for Autonomous Drivingcited by 68
  12. 2024P-MapNet: Far-seeing Map Generator Enhanced by both SDMap and HDMap Priorscited by 54
  13. 2022Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentationcited by 47
  14. 2022Patchwork++: Fast and Robust Ground Segmentation Solving Partial Under-Segmentation Using 3D Point Cloudcited by 47
  15. 2023Localization Is All You Evaluate: Data Leakage in Online Mapping Datasets and How to Fix Itcited by 44
  16. 2023Complementing Onboard Sensors with Satellite Map: A New Perspective for HD Map Constructioncited by 28
  17. 2019Learnable Tree Filter for Structure-preserving Feature Transformcited by 25
  18. 2022UniFusion: Unified Multi-view Fusion Transformer for Spatial-Temporal Representation in Bird's-Eye-Viewcited by 25
  19. 2023LiDAR2Map: In Defense of LiDAR-Based Semantic Map Construction Using Online Camera Distillationcited by 18
  20. 2024DiffMap: Enhancing Map Segmentation with Map Prior Using Diffusion Modelcited by 13
  21. 2024SemVecNet: Generalizable Vector Map Generation for Arbitrary Sensor Configurationscited by 12
  22. 2023BroadBEV: Collaborative LiDAR-camera Fusion for Broad-sighted Bird's Eye View Map Constructioncited by 4
  23. 2025MapKD: Unlocking Prior Knowledge with Cross-Modal Distillation for Efficient Online HD Map Constructioncited by 3
  24. 2026SatMap: Revisiting Satellite Maps as Prior for Online HD Map Constructioncited by 3
  25. 2020Laser Data Based Automatic Generation of Lane-Level Road Map for Intelligent Vehiclescited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025BridgeTA: Bridging the Representation Gap in Knowledge Distillation via Teacher Assistant for Bird's Eye View Map Segmentation
  2. 2025Domain Adaptation-Based Crossmodal Knowledge Distillation for 3D Semantic Segmentation
  3. 2023Knowledge Distillation from 3D to Bird's-Eye-View for LiDAR Semantic Segmentation
  4. 2023Rethinking Range View Representation for LiDAR Segmentation
  5. 2025The Devil is in the Details: Simple Remedies for Image-to-LiDAR Representation Learning
  6. 2024DiffSSC: Semantic LiDAR Scan Completion using Denoising Diffusion Probabilistic Models
  7. 20222DPASS: 2D Priors Assisted Semantic Segmentation on LiDAR Point Clouds
  8. 2023CMDFusion: Bidirectional Fusion Network with Cross-modality Knowledge Distillation for LIDAR Semantic Segmentation