CylinderDepth: Cylindrical Spatial Attention for Multi-View Consistent Self-Supervised Surround Depth Estimation
By: Samer Abualhanud, Christian Grannemann, Max Mehltretter
Potential Business Impact:
Makes 3D pictures from many cameras match perfectly.
Self-supervised surround-view depth estimation enables dense, low-cost 3D perception with a 360° field of view from multiple minimally overlapping images. Yet, most existing methods suffer from depth estimates that are inconsistent between overlapping images. Addressing this limitation, we propose a novel geometry-guided method for calibrated, time-synchronized multi-camera rigs that predicts dense, metric, and cross-view-consistent depth. Given the intrinsic and relative orientation parameters, a first depth map is predicted per image and the so-derived 3D points from all images are projected onto a shared unit cylinder, establishing neighborhood relations across different images. This produces a 2D position map for every image, where each pixel is assigned its projected position on the cylinder. Based on these position maps, we apply an explicit, non-learned spatial attention that aggregates features among pixels across images according to their distances on the cylinder, to predict a final depth map per image. Evaluated on the DDAD and nuScenes datasets, our approach improves the consistency of depth estimates across images and the overall depth compared to state-of-the-art methods.
Similar Papers
Depth-Consistent 3D Gaussian Splatting via Physical Defocus Modeling and Multi-View Geometric Supervision
CV and Pattern Recognition
Makes 3D pictures more real, near and far.
Omnidirectional Depth-Aided Occupancy Prediction based on Cylindrical Voxel for Autonomous Driving
CV and Pattern Recognition
Helps self-driving cars see in 3D better.
Integrating Disparity Confidence Estimation into Relative Depth Prior-Guided Unsupervised Stereo Matching
CV and Pattern Recognition
Helps robots see depth without needing labeled training data.