Depth completion is widely employed in Simultaneous Localization and Mapping (SLAM) and Structure from Motion (SfM). which are of great significance to the development of autonomous driving. Recently. the methods based on the fusion of vision transformer (ViT) and convolution have brought the accuracy to a new level. https://www.roneverhart.com/Clos-Pegase-Cabernet-Sauvignon-Hommage-Tenma-Vineyard-Napa-Valley-2014/