Cross-Vehicle 3D Geometric Consistency for Self-Supervised Surround Depth Estimation on Articulated Vehicles
This work addresses a specific challenge in autonomous driving for articulated vehicles, offering an incremental improvement over existing methods by adapting them to handle articulated structures.
The paper tackles the problem of surround depth estimation for articulated vehicles, which have complex geometry and motion, by proposing a self-supervised framework that uses cross-view and cross-vehicle geometric consistency, achieving state-of-the-art performance on multiple benchmarks including DDAD, nuScenes, and KITTI.
Surround depth estimation provides a cost-effective alternative to LiDAR for 3D perception in autonomous driving. While recent self-supervised methods explore multi-camera settings to improve scale awareness and scene coverage, they are primarily designed for passenger vehicles and rarely consider articulated vehicles or robotics platforms. The articulated structure introduces complex cross-segment geometry and motion coupling, making consistent depth reasoning across views more challenging. In this work, we propose \textbf{ArticuSurDepth}, a self-supervised framework for surround-view depth estimation on articulated vehicles that enhances depth learning through cross-view and cross-vehicle geometric consistency guided by structural priors from vision foundation model. Specifically, we introduce multi-view spatial context enrichment strategy and a cross-view surface normal constraint to improve structural coherence across spatial and temporal contexts. We further incorporate camera height regularization with ground plane-awareness to encourage metric depth estimation, together with cross-vehicle pose consistency that bridges motion estimation between articulated segments. To validate our proposed method, an articulated vehicle experiment platform was established with a dataset collected over it. Experiment results demonstrate state-of-the-art (SoTA) performance of depth estimation on our self-collected dataset as well as on DDAD, nuScenes, and KITTI benchmarks.