Three-dimensional point cloud generation using a polarimetric camera in a drive assistance system equipped vehicle

A method performed by a processor comprising:

• receiving a color image of a field of view from a camera;
• receiving a polarimetric data of the field of view from a polarimetric camera;
• performing a neural network function to generate a depth map of the field of view in response to the color image and the polarimetric data;
• and
• performing a vehicle control operation to control a vehicle in response to the depth map.

A method for performing an assisted driving operation comprising:

• receiving a first polarimetric data of a first field of view;
• receiving a first color image of the first field of view;
• receiving a lidar depth point cloud of the first field of view;
• training a neural network function in response to the first polarimetric data, the first color image, and the lidar depth point cloud;
• receiving a second color image of a second field of view and a second polarimetric data of the second field of view;
• performing the neural network function to generate a depth map of the second field of view in response to the second color image and the second polarimetric data;
• and
• performing the assisted driving operation in response to the depth map to control a vehicle.