System and method for object recognition using depth mapping
AssigneeChiral Software, Inc.
InventorsEric Jonathan Hollander, Michael Travis Remington
Life of this patent
Estimating the term…
Abstract
A method for performing real-time recognition of objects in motion includes receiving an input video stream from a camera, generating one or more depth maps for one or more frames of the input video stream, recognizing one or more objects in a current frame based on corresponding depth map using a machine learning algorithm, and displaying the one or more recognized objects in the current frame in one or more bounding boxes.
Claims24
A method for performing real-time recognition of objects, the method comprising:
- receiving an input video stream from a camera;
- generating one or more depth maps for one or more frames of the input video stream;
- recognizing one or more objects in a current frame based on a corresponding depth map, using a machine learning algorithm, wherein recognition of the one or more objects is triggered when the one or more objects appear at a depth less than or equal to a predefined depth in the corresponding depth map;
- and
- displaying the one or more recognized objects in the current frame in one or more bounding boxes.
A system for performing real-time recognition of objects, the system comprising:
- a camera to generate an input video stream;
- a depth mapping module configured to:
- receive the input video stream from the camera;
- generate one or more depth maps for one or more frames of the input video stream;
- an object recognizer configured to recognize one or more objects in a current frame based on a corresponding depth map, using a machine learning algorithm, wherein recognition of the one or more objects is triggered when the one or more objects appear at a depth less than or equal to a predefined depth in the corresponding depth map;
- and
- a display device configured to display the one or more recognized objects in the current frame in one or more bounding boxes.
An unmanned aerial vehicle (UAV) camera system for a UAV, comprising:
- a camera;
- a display device;
- a memory to store one or more instructions;
- a processor communicatively coupled to the camera, display, and the memory, wherein the processor is configured to execute the one or more instructions of the memory to:
- receive an input video stream from a camera while the UAV is in motion;
- generate one or more depth maps for one or more frames of the input video stream using parallax depth mapping;
- recognize one or more objects in a current frame based on a corresponding depth map, using a machine learning algorithm, wherein recognition of the one or more objects is triggered when the one or more objects appear at a depth less than or equal to a predefined depth in the corresponding depth map;
- and
- display the one or more recognized objects in the current frame in one or more bounding boxes.
A computer programmable product for performing real-time recognition of objects, the computer programmable product comprising a set of instructions, the set of instructions when executed by a processor causes the processor to:
- receive an input video stream from a camera;
- generate one or more depth maps for one or more frames of the input video stream;
- recognize one or more objects in a current frame based on a corresponding depth map, using a machine learning algorithm, wherein recognition of the one or more objects is triggered when the one or more objects appear at a depth less than or equal to a predefined depth in the corresponding depth map;
- and
- display the one or more recognized objects in the current frame in one or more bounding boxes.
Dependencies are inferred from the claim text.