Systems and methods for depth estimation using monocular images

A non-transitory computer-readable medium for generating depth estimates from a monocular image and including instructions that when executed by one or more processors cause the one or more processors to:

• flip, using a disparity model, the monocular image to generate a flipped image, wherein the disparity model is a machine learning algorithm;
• analyze, using the disparity model, the monocular image and the flipped image to generate disparity maps including a monocular disparity map corresponding to the monocular image and a flipped disparity map corresponding with the flipped image;
• generate, using the disparity model, a fused disparity map from the monocular disparity map and the flipped disparity map;
• wherein the instructions to generate the fused disparity map includes combining pixels between the monocular map and a re-flipped map to provide a combined output, the re-flipped map is horizontally flipped from the flipped disparity map;
• and
• provide the fused disparity map as the depth estimates of objects represented in the monocular image.

A method of generating depth estimates from a monocular image, comprising:

• in response to receiving the monocular image, flipping, by a disparity model, the monocular image to generate a flipped image, wherein the disparity model is a machine learning algorithm;
• analyzing, using the disparity model, the monocular image and the flipped image to generate disparity maps including a monocular disparity map corresponding to the monocular image and a flipped disparity map corresponding with the flipped image;
• generating, in the disparity model, a fused disparity map from the monocular disparity map and the flipped disparity map, wherein generating the fused disparity map includes combining pixels between the monocular map and a re-flipped map to provide a combined output, the re-flipped map is horizontally flipped from the flipped disparity map;
• and
• providing the fused disparity map as the depth estimates of objects represented in the monocular image.