MDT3D: Multi-Dataset Training for LiDAR 3D Object Detection Generalization
This addresses the real-world issue of deploying LiDAR-based detection systems in new environments without target data, though it is incremental as it builds on existing multi-dataset approaches.
The paper tackles the problem of 3D object detection models failing to generalize to unseen datasets with different sensor configurations by proposing MDT3D, a multi-dataset training method that improves robustness through label mapping and cross-dataset augmentation, showing improvements across different model types.
Supervised 3D Object Detection models have been displaying increasingly better performance in single-domain cases where the training data comes from the same environment and sensor as the testing data. However, in real-world scenarios data from the target domain may not be available for finetuning or for domain adaptation methods. Indeed, 3D object detection models trained on a source dataset with a specific point distribution have shown difficulties in generalizing to unseen datasets. Therefore, we decided to leverage the information available from several annotated source datasets with our Multi-Dataset Training for 3D Object Detection (MDT3D) method to increase the robustness of 3D object detection models when tested in a new environment with a different sensor configuration. To tackle the labelling gap between datasets, we used a new label mapping based on coarse labels. Furthermore, we show how we managed the mix of datasets during training and finally introduce a new cross-dataset augmentation method: cross-dataset object injection. We demonstrate that this training paradigm shows improvements for different types of 3D object detection models. The source code and additional results for this research project will be publicly available on GitHub for interested parties to access and utilize: https://github.com/LouisSF/MDT3D