REAL-TIME CASCADED OBJECT RECOGNITION

• US 20170091953A1
• Filed: 09/25/2015
• Published: 03/30/2017
• Est. Priority Date: 09/25/2015
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Amir Rosenberger
• Inventors: Oshrat, Yinon, Bleiweiss, Amit, Yanai, Yaron, Rosenberger, Amir
• Granted Patent: US 9,928,605 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06T 1/60 Memory management
• CPC Class Codes: G06T 7/11 Region-based segmentation
• CPC Class Codes: G06T 7/13 Edge detection
• CPC Class Codes: G06T 7/246 using feature-based methods, e.g. the tracking of corners or segments
• CPC Class Codes: G06V 20/52 Surveillance or monitoring of activities, e.g. for recognising suspicious objects recognising microscopic objects G06V20/69
• CPC Class Codes: G06V 40/174 Facial expression recognition
• CPC Class Codes: G06V 40/178 estimating age from face image; using age information for improving recognition
• CPC Class Codes: H04N 13/204 using stereoscopic image cameras stereoscopic photography G03B35/00
• CPC Class Codes: H04N 13/271 wherein the generated image signals comprise depth maps or disparity maps
• CPC Class Codes: H04N 5/33 Transforming infrared radiation cameras or camera modules for generating image signals from infrared radiation H04N23/20; circuitry of SSIS for transforming infrared radiation into image signals H04N25/20

Abstract

Various systems and methods for real-time cascaded object recognition are described herein. A system for real-time cascaded object recognition comprises a processor; and a memory, including instructions, which when executed on the processor, cause the processor to perform the operations comprising: accessing image data at the system, the image data of an environment around the system, the image data is captured by a camera system; determining a set of regions in the image data, the set of regions including candidate objects; transmitting a subset of the image data corresponding to the set of regions to a remote server, the remote server to analyze the subset of the image data and detect an object in the subset of the image data; and receiving at the system from the remote server, an indication of the object detected in the subset of the image data.

Claims

1. A system for real-time cascaded object recognition, the system comprising:

a processor; and

a memory, including instructions, which when executed on the processor, cause the processor to perform the operations comprising;

accessing image data at the system, the image data of an environment around the system, the image data is captured by a camera system;

determining a set of regions in the image data, the set of regions including candidate objects;

transmitting a subset of the image data corresponding to the set of regions to a remote server, the remote server to analyze the subset of the image data and detect an object in the subset of the image data; and

receiving at the system from the remote server, an indication of the object detected in the subset of the image data.

2. The system of claim 1, wherein the processor is to further perform operations comprising determining the set of regions in the image data comprises using an image segmentation process.

3. The system of claim 2, wherein the image segmentation process comprises a depth segmentation process.

4. A system for real-time cascaded object recognition, the system comprising:

an access module to access image data at the system, the image data of an environment around the system, the image data is captured by a camera system;

a candidate determination module to determine a set of regions in the image data, the set of regions including candidate objects;

a transmission module to transmit a subset of the image data corresponding to the set of regions to a remote server, the remote server to analyze the subset of the image data and detect an object in the subset of the image data; and

a receive module to receive at the system from the remote server, an indication of the object detected in the subset of the image data.

5. The system of claim 4, wherein the system comprises an image compression module to compress the subset of the image data before transmitting the subset of the image data to the remote server.

6. The system of claim 4, wherein the camera system is incorporated into the system.

7. The system of claim 4, wherein the camera system comprises a depth camera.

8. The system of claim 4, wherein the candidate determination module to determine the set of regions in the image data using an image segmentation process.

9. The system of claim 8, wherein the image segmentation process comprises a depth segmentation process.

10. The system of claim 8, wherein the image segmentation process comprises an edge detection segmentation process.

11. The system of claim 4, wherein the camera system comprises a depth camera and the subset of image data comprises pixels less than a threshold distance away from the camera system as measured by the depth camera.

12. At least one machine-readable medium including instructions, which when executed by a machine, cause the machine to:

access image data at a compute device, the image data of an environment around the compute device, the image data is captured by a camera system;

determine a set of regions in the image data, the set of regions including candidate objects;

transmit a subset of the image data corresponding to the set of regions to a remote server, the remote server to analyze the subset of the image data and detect an object in the subset of the image data; and

receive at the compute device from the remote server, an indication of the object detected in the subset of the image data.

13. The at least one machine-readable medium of claim 12, further comprising instructions to compress the subset of the image data before transmitting the subset of the image data to the remote server.

14. The at least one machine-readable medium of claim 12, wherein the camera system comprises a depth camera and the subset of image data comprises pixels less than a threshold distance away from the camera system as measured by the depth camera.

15. The at least one machine-readable medium of claim 12, wherein the remote server analyzes the subset of the image data using a machine learning process.

16. The at least one machine-readable medium of claim 12, wherein the indication of the object comprises an instruction.

17. The at least one machine-readable medium of claim 16, wherein the instruction comprises an instruction to the compute device to maneuver to a different position to capture additional image data to supplement the image data.

18. The at least one machine-readable medium of claim 12, further comprising instructions to display the indication of the object on a display of the compute device.

19. A method of real-time cascaded object recognition, the method comprising:

accessing image data at a compute device, the image data of an environment around the compute device, the image data is captured by a camera system;

determining a set of regions in the image data, the set of regions including candidate objects;

transmitting a subset of the image data corresponding to the set of regions to a remote server, the remote server to analyze the subset of the image data and detect an object in the subset of the image data; and

receiving at the compute device from the remote server, an indication of the object detected in the subset of the image data.

20. The method of claim 19, further comprising compressing the subset of the image data before transmitting the subset of the image data to the remote server.

21. The method of claim 19, wherein the camera system comprises a depth camera and the subset of image data comprises pixels less than a threshold distance away from the camera system as measured by the depth camera.

22. The method of claim 19, wherein the remote server analyzes the subset of the image data using a machine learning process.

23. The method of claim 19, wherein the indication of the object comprises an instruction.

24. The method of claim 23, wherein the instruction comprises an instruction to the compute device to maneuver to a different position to capture additional image data to supplement the image data.

25. The method of claim 19, further comprising displaying the indication of the object on a display of the compute device.