When used by autonomous vehicles for trajectory planning or obstacle avoidance, depth estimation methods need to be reliable. Therefore, estimating the quality of the depth outputs is critical. In this paper, we show how M4Depth, a state-of-the-art depth estimation method designed for unmanned aerial vehicle (UAV) applications, can be enhanced to perform joint depth and uncertainty estimation. For that, we present a solution to convert the uncertainty estimates related to parallax generated by M4Depth into uncertainty estimates related to depth, and show that it outperforms the standard probabilistic approach. Our experiments on various public datasets demonstrate that our method performs consistently, even in zero-shot transfer. Besides, our method offers a compelling value when compared to existing multi-view depth estimation methods as it performs similarly on a multi-view depth estimation benchmark despite being 2.5 times faster and causal, as opposed to other methods. The code of our method is publicly available at https://github.com/michael-fonder/M4DepthU .

翻译：当自主车辆用于轨迹规划或障碍物规避时，深度估计方法必须具有可靠性。因此，评估深度输出的质量至关重要。本文展示了如何增强M4Depth（一种专为无人机应用设计的最先进深度估计方法），使其能够执行联合深度与不确定性估计。为此，我们提出一种解决方案，将M4Depth生成的与视差相关的不确定性估计转换为与深度相关的不确定性估计，并证明其优于标准概率方法。我们在多个公开数据集上的实验表明，该方法即使在零样本迁移场景下也能保持一致性。此外，与现有多视图深度估计方法相比，我们的方法展现出显著优势：尽管速度提高了2.5倍且具有因果性，但在多视图深度估计基准测试中性能相当。我们的方法代码已公开在https://github.com/michael-fonder/M4DepthU。