Leveraging multidimensional sensor data for computationally efficient object detection for autonomous machine applications

A system comprising:

• one or more processors to:
• determine, based at least on sensor data generated using one or more first sensors of a machine, a two-dimensional (2D) bounding shape associated with an object;
• project the 2D bounding shape from a 2D sensor representation corresponding to the sensor data to a 3D sensor representation corresponding to depth data generated using one or more second sensors of the machine;
• determine, based at least on the projection, a portion of the depth data that is associated with the object;
• and
• cause, based at least on the portion of the depth data, the machine to perform one or more operations.

A processor comprising:

• processing circuitry to:
• determine, based at least on first sensor data generated using one or more first sensors of a machine, a first two-dimensional (2D) bounding shape associated with an object;
• generate, based at least on second sensor data generated using one or more second sensors of the machine, a depth image;
• determine, based at least on the first 2D bounding shape, a second 2D bounding shape associated with the depth image;
• determine, based at least on the second 2D bounding shape, one or more points associated with the depth image;
• and
• determine, based at least on the one or more points, a 3D bounding shape associated with the object.