Object detection techniques

• US 9,589,179 B2
• Filed: 12/19/2013
• Issued: 03/07/2017
• Est. Priority Date: 12/19/2013
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Nielsen, Andrew Sherman, Wu, Yongjun
• Primary Examiner(s): Seth, Manav
• Application Number:
  US14/133,814
• Publication Number:
  US 20150178552A1
• Time in Patent Office: 1,174 Days
• US Class Current: 1/1
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/30201 Face
• CPC Class Codes: G06T 7/20 Analysis of motion motion estimation for coding, decoding, compressing or decompressing digital video signals H04N19/43, H04N19/51
• CPC Class Codes: G06V 10/759 Region-based matching
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 30/2504 Coarse or fine approaches, e.g. resolution of ambiguities or multiscale approaches
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/167 using comparisons between temporally consecutive images
• CPC Class Codes: G06V 40/20 Movements or behaviour, e.g. gesture recognition recognition of facial expressions G06V40/16
• CPC Class Codes: G08B 13/19602 Image analysis to detect motion of the intruder, e.g. by frame subtraction
• CPC Class Codes: G08B 13/1961 Movement detection not involving frame subtraction, e.g. motion detection on the basis of luminance changes in the image

Abstract

Object detection techniques are described. In one or more implementations, a plurality of images are received by a computing device. The plurality of images are analyzed by the computing device to detect whether the images include, respectively, a depiction of an object. If an object is found in a first image, the locations, angles and scales for object detection can be further restricted in a second one. If an object is not found in a first one of the image, different portions of a second one of the images are analyzed for object detection.

Claims

1. A method comprising:

receiving a plurality of images by a computing device, the computing device having a processor including a first processor core and a second processor core;

determining, by the computing device, whether an object has been previously detected in one or more frames of the plurality of images;

responsive to a determination that the object has not been detected, employing the first processor core and the second processor core of the computing device for analyzing one or more subsequent frames of the plurality of images by the computing device to detect whether the images include, respectively, a depiction of the object, the analyzing performed such that one or more portions of the plurality of images used to detect the object by the first processor core is different than one or more portions of the plurality of images used to detect the object by the second processor core; and

presenting object detection information for the object to a user via an output device.

2. A method as described in claim 1, wherein the difference in the one or more portions for the respective first and second said images involves a scale assigned for the object to perform the analyzing.

3. A method as described in claim 1, wherein the difference in the one or more portions for the respective first and second said images involves a region of the respective first and second said images that is utilized to perform the analyzing.

4. A method as described in claim 1, wherein the difference in the one or more portions for the respective first and second said images involves an angle at which the object is to be detected in the respective first and second said images.

5. A method as described in claim 1, wherein the difference in the one or more portions for at least one of the respective first and second said images is based on a previous detection of the object in one or more of the plurality of frames.

6. A method as described in claim 5, wherein the previous detection in indicates a scale of the object that is to be used as a basis to perform the analysis.

7. A method as described in claim 5, wherein the previous detection in indicates a location of the object in the one or more of the plurality of images that is to be used as a basis to perform the analysis.

8. A method as described in claim 5, wherein the previous detection in indicates motion of the object in the one or more of the plurality of images that is to be used as a basis to perform the analysis.

9. A method as described in claim 8, wherein the motion indicates a direction of movement of the object in the one or more of the plurality of images.

10. A method as described in claim 5, wherein the previous detection in indicates an angle of the object in the one or more of the plurality of images that is to be used as a basis to perform the analysis.

11. A method as described in claim 1, wherein the object is a human face.

12. A method as described in claim 1, wherein the analyzing is configured to be performed such that detection of the object by one of the processor cores causes operations performed by the other processor core to perform the analyzing to cease.

13. A method as described in claim 1, wherein the analyzing is performed by the first processor core and the second processor core in parallel.

14. A system comprising:

one or more processors; and

one or more computer-readable storage media storing instructions that are executable by the one or more processors to perform operations including;

receiving a plurality of images by a computing device, the computing device having a processor including a first processor core and a second processor core;

determining, by the computing device, whether an object has been previously detected in one or more frames of the plurality of images;

responsive to a determination that the object has not been detected, employing the first processor core and the second processor core of the computing device for analyzing one or more subsequent frames of the plurality of images by the computing device to detect whether the images include, respectively, a depiction of the object, the analyzing performed such that one or more portions of the plurality of images used to detect the object by the first processor core is different than one or more portions of the plurality of images used to detect the object by the second processor core; and

presenting object detection information for the object to a user via an output device.

15. A system as described in claim 14, wherein the difference in the one or more portions for the respective first and second said images involves a scale assigned for the object to perform the analyzing.

16. A system as described in claim 14, wherein the difference in the one or more portions for the respective first and second said images involves a region of the respective first and second said images that is utilized to perform the analyzing.

17. A system as described in claim 14, wherein the difference in the one or more portions for the respective first and second said images involves an angle at which the object is detected.

18. A system as described in claim 14, wherein the analyzing is performed by the first processor core and the second processor core in parallel.

19. A system as described in claim 14, wherein the analyzing is configured to be performed such that detection of the object by one of the processor cores causes operations performed by the other processor core to perform the analyzing to cease.

20. A system as described in claim 14, wherein the object is a human face.

21. A method comprising:

determining whether an object has been previously detected in one or more images;

responsive to a determination that the object has not been detected, employing a processor with a first processor core and a second processor core for analyzing the one or more images to detect a depiction of the object;

separating execution of the analyzing on the first processor core and the second processor core such that the first processor core and the second processor core utilize different respective portions of the one or more images to perform the analyzing to detect the object; and

presenting the object detection information for the object to a user via an output device.

22. A method as described in claim 21, wherein the difference in the different portions involves a scale assigned for the object to perform the analyzing.

23. A method as described in claim 21, wherein the difference in the different portions for the respective images involves a region of the respective images that is utilized to perform the analyzing.

24. A method as described in claim 21, wherein the difference in the different portions for the respective images involves an angle at which the object is detected.

25. A method as described in claim 21, wherein the object is a human face.

26. A method as described in claim 21, wherein the analyzing is performed by the first processor core and the second processor core in parallel.

27. A method as described in claim 21, wherein the analyzing is configured to be performed such that detection of the object by one of the processor cores causes operations performed by the other processor core to perform the analyzing to cease.