Uncertainty depth estimation with gated images for 3D reconstruction
This work addresses the need for reliable depth estimation in autonomous vehicles under adverse weather conditions, but it is incremental as it builds upon an existing framework.
The paper tackled the problem of improving depth estimation accuracy for 3D reconstruction in self-driving cars by extending the Gated2Depth framework to include aleatoric uncertainty, which provides confidence measures to filter out uncertain depth estimates, particularly in poorly illuminated regions, and showed that training on dense depth maps from LiDAR depth completion algorithms further enhances performance.
Gated imaging is an emerging sensor technology for self-driving cars that provides high-contrast images even under adverse weather influence. It has been shown that this technology can even generate high-fidelity dense depth maps with accuracy comparable to scanning LiDAR systems. In this work, we extend the recent Gated2Depth framework with aleatoric uncertainty providing an additional confidence measure for the depth estimates. This confidence can help to filter out uncertain estimations in regions without any illumination. Moreover, we show that training on dense depth maps generated by LiDAR depth completion algorithms can further improve the performance.