Structured Domain Randomization: Bridging the Reality Gap by Context-Aware Synthetic Data
This addresses the problem of bridging the reality gap for computer vision researchers and practitioners by generating more realistic synthetic training data, though it is incremental as it builds on domain randomization.
The paper tackles the reality gap in synthetic data training for object detection by introducing structured domain randomization (SDR), which places objects and distractors based on context-aware probability distributions instead of uniform randomness. It demonstrates competitive results on real KITTI data after training only on synthetic data, outperforming other synthetic and real data methods, with SDR combined with real data further improving performance.
We present structured domain randomization (SDR), a variant of domain randomization (DR) that takes into account the structure and context of the scene. In contrast to DR, which places objects and distractors randomly according to a uniform probability distribution, SDR places objects and distractors randomly according to probability distributions that arise from the specific problem at hand. In this manner, SDR-generated imagery enables the neural network to take the context around an object into consideration during detection. We demonstrate the power of SDR for the problem of 2D bounding box car detection, achieving competitive results on real data after training only on synthetic data. On the KITTI easy, moderate, and hard tasks, we show that SDR outperforms other approaches to generating synthetic data (VKITTI, Sim 200k, or DR), as well as real data collected in a different domain (BDD100K). Moreover, synthetic SDR data combined with real KITTI data outperforms real KITTI data alone.