mlsanityData to 14 Oct 2026

Synth It Like KITTI: Synthetic Data Generation for Object Detection in Driving Scenarios

arXiv 2502.15076 ↗First posted 20 Feb 2025cs.CV cs.ROcited by 3

Not on the map. Why?

Abstract (arXiv)

An important factor in advancing autonomous driving systems is simulation. Yet, there is rather small progress for transferability between the virtual and real world. We revisit this problem for 3D object detection on LiDAR point clouds and propose a dataset generation pipeline based on the CARLA simulator. Utilizing domain randomization strategies and careful modeling, we are able to train an object detector on the synthetic data and demonstrate strong generalization capabilities to the KITTI dataset. Furthermore, we compare different virtual sensor variants to gather insights, which sensor attributes can be responsible for the prevalent domain gap. Finally, fine-tuning with a small portion of real data almost matches the baseline and with the full training set slightly surpasses it.

Key facts. Cited by 3 papers in the map (data through 14 Oct 2026). Builds on 19 of its 32 references that are in the map. First posted on arXiv 20 Feb 2025.

Builds on

19 of 32 references.

References · 19

  1. 2019nuScenes: A multimodal dataset for autonomous drivingcited by 4,523
  2. 2019Scalability in Perception for Autonomous Driving: Waymo Open Datasetcited by 2,451
  3. 2017Domain Randomization for Transferring Deep Neural Networks from Simulation to the Real Worldcited by 1,498
  4. 2023Argoverse 2: Next Generation Datasets for Self-Driving Perception and Forecastingcited by 707
  5. 2017AirSim: High-Fidelity Visual and Physical Simulation for Autonomous Vehiclescited by 523
  6. 2020Voxel R-CNN: Towards High Performance Voxel-based 3D Object Detectioncited by 410
  7. 2016Virtual Worlds as Proxy for Multi-Object Tracking Analysiscited by 341
  8. 2017Joint 3D Proposal Generation and Object Detection from View Aggregationcited by 326
  9. 2021One Million Scenes for Autonomous Driving: ONCE Datasetcited by 218
  10. 2020Train in Germany, Test in The USA: Making 3D Object Detectors Generalizecited by 151
  11. 2021ST3D++: Denoised Self-training for Unsupervised Domain Adaptation on 3D Object Detectioncited by 57
  12. 2023Uni3D: A Unified Baseline for Multi-dataset 3D Object Detectioncited by 39
7 more
  1. 2021Paris-CARLA-3D: A Real and Synthetic Outdoor Point Cloud Dataset for Challenging Tasks in 3D Mappingcited by 25
  2. 2023AmodalSynthDrive: A Synthetic Amodal Perception Dataset for Autonomous Drivingcited by 14
  3. 2023Quantifying the LiDAR Sim-to-Real Domain Shift: A Detailed Investigation Using Object Detectors and Analyzing Point Clouds at Target-Levelcited by 7
  4. 2022A Realism Metric for Generated LiDAR Point Cloudscited by 5
  5. 2023ParisLuco3D: A high-quality target dataset for domain generalization of LiDAR perceptioncited by 5
  6. 2019Multimodal 3D Object Detection from Simulated Pretrainingcited by 4
  7. 2023GAN-Based LiDAR Intensity Simulationcited by 1

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 1; 2026: 2.

3 papers.

  1. 2025Physically Based Neural LiDAR Resimulationcited by 1
  2. 2026VERIA: Verification-Centric Multimodal Instance Augmentation for Long-Tailed 3D Object Detectioncited by 0
  3. 2026Object Detection for Autonomous Driving in Chinese Rural Scenes: An Experimental Study on Real-Synthetic Data Mixing and Model Evaluationcited by 0