BRIDGE -- Building Reinforcement-Learning Depth-to-Image Data Generation Engine for Monocular Depth Estimation

Fudan University Figure 1: We present BRIDGE, showcasing its RL-optimized Depth-to-Image (D2I) data generation engine which is used for generating realistic and geometrically accurate RGB images from source depth maps and Monocular Depth Estimation (MDE) model which after being trained on the massive high-quality data generated by the D2I engine, achieves superior depth prediction in complex scenes. Monocular Depth Estimation (MDE) is a foundational task for computer vision. Traditional methods are limited by data scarcity and quality, hindering their robustness. To overcome this, we propose BRIDGE, an RL-optimized depth-to-image (D2I) generation framework that synthesizes over 20M realistic and geometrically accurate RGB images, each intrinsically paired with its ground truth depth, from diverse source depth maps. Then we train our depth estimation model on this dataset, employing a hybrid supervision strategy that integrates teacher pseudo-labels with ground truth depth for comprehensive and robust training. This innovative data generation and training paradigm enables BRIDGE to achieve breakthroughs in scale and domain diversity, consistently outperforming existing state-of-the-art approaches quantitatively and in complex scene detail capture, thereby fostering general and robust depth features. Existing data acquisition and utilization methods can be categorized into three groups. While valuable, this approach is constrained by the sparsity of depth maps from sensors and the difficulty of accurately capturing and annotating transparent or reflective objects (Costanzino et al., 2023).