System and method for estimating depth using a monocular camera

A system for estimating depth using a monocular camera, the system comprising:

• one or more processors;
• a monocular camera operably connected to the one or more processors;
• a memory device operably connected to the one or more processors, the memory device having an image capture module having instructions that when executed by the one or more processors cause the one or more processors to obtain a captured image from the monocular camera;
• the memory device further comprising a style transfer module, the style transfer module having instructions that when executed by the one or more processors cause the one or more processors to generate a synthesized image based on the captured image, wherein the style transfer module was trained using a generative adversarial network;
• the memory device further comprising an encoder-decoder module, the encoder-decoder module having instructions that when executed by the one or more processors cause the one or more processors to generate a feature map based on the synthesized image, wherein the feature map includes at least one matrix having at least one numerical element;
• the memory device further comprising a semantic information generating module, the semantic information generating module having instructions that when executed by the one or more processors cause the one or more processors to generate semantic information based on the feature map, the semantic information indicating a boundary for at least one structure in the captured image;
• and
• the memory device further comprising a depth map generating module, the depth map generating module having instructions that when executed by the one or more processors cause the one or more processors to generate a depth map based on the feature map and the semantic information, wherein the depth map includes an estimated depth for at least a portion of pixels of the captured image.