BRIDGE -- Building Reinforcement-Learning Depth-to-Image Data Generation Engine for Monocular Depth Estimation
This addresses data limitations for computer vision researchers and practitioners in MDE, offering a novel data generation paradigm that is not incremental but introduces a new method for a known bottleneck.
The paper tackles the problem of data scarcity and quality in Monocular Depth Estimation (MDE) by proposing BRIDGE, an RL-optimized depth-to-image generation framework that synthesizes over 20M realistic RGB images with ground truth depth, leading to breakthroughs in scale and domain diversity and consistently outperforming existing state-of-the-art approaches.
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. Code and models are available at https://dingning-liu.github.io/bridge.github.io/.