Processing sensor information for object detection

A vehicle control unit, comprising:

• a processor configured with processor-executable instructions to perform operations comprising:
• receiving sensor information from a sensor;
• generating a feature map based on the sensor information;
• associating with each pixel of the feature map a prior box comprising:
• a plurality of width priors that each include a width value,
• a plurality of height priors that each include a height value, and
• a plurality of orientation priors that each include an orientation value;
• determining a confidence value of each height prior of the plurality of height priors, each width prior of the plurality of width priors for each prior box, and each orientation prior of the plurality of orientation priors;
• outputting an indication of a detected object based on a highest confidence height prior from among the plurality of height priors a highest confidence width prior from among the plurality of width priors, and a highest confidence orientation prior from among the plurality of orientation priors;
• and
• taking an action based on the output indication of a detected object.

A method of processing vehicle sensor information for object detection, comprising:

• capturing, by a sensor, sensor information regarding an environment around the sensor;
• generating, by a processor, a feature map based on the sensor information;
• associating, by the processor, with each pixel of the feature map, one or more prior boxes each comprising a center value, a size value, and a plurality of orientation priors that each include an orientation value;
• determining, by the processor, a confidence value for each orientation prior of the plurality of orientation priors;
• outputting, by the processor, an indication of a detected object based on the highest confidence orientation;
• and
• performing, by the processor, a vehicle operation based on the output indication of a detected object.

A vehicle control unit, comprising:

• a processor configured with processor-executable instructions to perform operations comprising:
• receiving sensor information regarding an environment around a sensor;
• generating, by a processor, a feature map based on the sensor information;
• associating, by the processor, with each pixel of the feature map, one or more prior boxes each comprising a center value, a size value, and a plurality of orientation priors that each include an orientation value;
• determining, by the processor, a confidence value for each orientation prior of the plurality of orientation priors;
• outputting, by the processor, an indication of a detected object based on the highest confidence orientation;
• and
• performing, by the processor, a vehicle operation based on the output indication of a detected object.