Fig. 10From: An image-guided network for depth edge enhancementVisualization results on Flyingthings3D sub-dataset: a color map, b ground-truth depth, c depth map with errors, d refined depth map, e error map before refinement, f error map after refinementBack to article page