Self-Supervised Monocular Depth Estimation With Depth–Motion Prior for Pseudo-LiDAR
Shuangjie Yuan, Houxuan Liu, Yipeng Liu, Lu Yang · IEEE Transactions on Instrumentation and Measurement · 2025
In recent years, learning-based depth estimation has advanced significantly due to multi-view geometry and self-supervised methods. These approaches use video sequences or multi-view images for supervision, making training data easier to obtain. However, challenges remain, including the need for extrinsic camera pre-calibration and insufficient real-time performance. To address these issues, we propose DMPDepth, a self-supervised monocular depth estimation method with multi-view stereo prior. This approach integrates the computational efficiency of monocular depth estimation with the accuracy of multi-view stereo matching. We incorporate a pose estimation module that eliminates the need for camera extrinsic pre-calibration. In addition, a depth motion prior is introduced by combining triangulation with camera motion, and a depth motion prior guidance strategy is proposed to improve real-time performance and accuracy. We demonstrate the significantly superior performance of DMPDepth over other public methods on the KITTI and self-constructed Campus Road datasets, with reduced inference time. Furthermore, a pseudo-LiDAR system for autonomous driving based on DMPDepth successfully demonstrates accurate real-time depth estimation in a real Campus Road environment.