Deep Multi-view Depth Estimation with Predicted Uncertainty
Deep Multi-view Depth Estimation with Predicted Uncertainty
复制标题
DOI：
10.1109/icra48506.2021.9560873
复制
发表时间：
2020-11
期刊：
影响因子：
--
通讯作者：
Tong Ke;Tien Do;Khiem Vuong;K. Sartipi;S. Roumeliotis
中科院分区：
文献类型：
--
作者：
Tong Ke;Tien Do;Khiem Vuong;K. Sartipi;S. Roumeliotis
在本文中，我们解决了使用深度神经网络从图像序列中估计密集深度的问题。具体来说，我们采用密集光流网络来计算对应关系，然后对点云进行三角测量以获得初始深度图。然而，由于缺乏共同观测或视差小，点云的部分可能不如其他部分准确。为了进一步提高三角测量的准确性，我们引入了一个深度细化网络（DRN），它可以根据图像的上下文线索优化初始深度图。特别地，DRN包含迭代细化模块（Iterative Refinement Module，缩写为ERF），其通过细化深度特征来提高迭代的深度精度。最后，DRN还预测细化深度的不确定性，这在诸如场景重建的测量选择的应用中是期望的。我们的实验表明，我们的算法在深度精度方面优于最先进的方法，并验证我们预测的不确定性与实际深度误差高度相关。
In this paper, we address the problem of estimating dense depth from a sequence of images using deep neural networks. Specifically, we employ a dense-optical-flow network to compute correspondences and then triangulate the point cloud to obtain an initial depth map. Parts of the point cloud, however, may be less accurate than others due to lack of common observations or small parallax. To further increase the triangulation accuracy, we introduce a depth-refinement network (DRN) that optimizes the initial depth map based on the image’s contextual cues. In particular, the DRN contains an iterative refinement module (IRM) that improves the depth accuracy over iterations by refining the deep features. Lastly, the DRN also predicts the uncertainty in the refined depths, which is desirable in applications such as measurement selection for scene reconstruction. We show experimentally that our algorithm outperforms state-of-the-art approaches in terms of depth accuracy, and verify that our predicted uncertainty is highly correlated to the actual depth error.