Cycle and Semantic Consistent Adversarial Domain Adaptation for Reducing Simulation-to-Real Domain Shift in LiDAR Bird's Eye View
This addresses the challenge of using synthetic data for training LiDAR-based object detection models, particularly for smaller road agents like pedestrians, though it is incremental as it builds on existing domain adaptation techniques.
The paper tackles the simulation-to-real domain shift problem in LiDAR bird's eye view object detection by proposing a CycleGAN-based domain adaptation method that uses semantic classification to preserve small object information, achieving improved results on the KITTI benchmark.
The performance of object detection methods based on LiDAR information is heavily impacted by the availability of training data, usually limited to certain laser devices. As a result, the use of synthetic data is becoming popular when training neural network models, as both sensor specifications and driving scenarios can be generated ad-hoc. However, bridging the gap between virtual and real environments is still an open challenge, as current simulators cannot completely mimic real LiDAR operation. To tackle this issue, domain adaptation strategies are usually applied, obtaining remarkable results on vehicle detection when applied to range view (RV) and bird's eye view (BEV) projections while failing for smaller road agents. In this paper, we present a BEV domain adaptation method based on CycleGAN that uses prior semantic classification in order to preserve the information of small objects of interest during the domain adaptation process. The quality of the generated BEVs has been evaluated using a state-of-the-art 3D object detection framework at KITTI 3D Object Detection Benchmark. The obtained results show the advantages of the proposed method over the existing alternatives.