CONTROLLING POWER CONSUMPTION IN OBJECT TRACKING PIPELINE

• US 20130141597A1
• Filed: 12/06/2011
• Published: 06/06/2013
• Est. Priority Date: 12/06/2011
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Inventors: Lee, Kyungsuk David, Finocchio, Mark, Moore, Richard, Balan, Alexandru, Fleck, Rod G.
• Granted Patent: US 9,171,380 B2
• Time in Patent Office: Days
• US Class Current: 348/207.11
• CPC Class Codes: G06T 2207/10016 Video; Image sequence
• CPC Class Codes: G06T 2207/20224 Image subtraction
• CPC Class Codes: G06T 2207/30196 Human being; Person
• CPC Class Codes: G06T 7/194 involving foreground-background segmentation
• CPC Class Codes: G06T 7/251 involving models
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: H04N 23/611 where the recognised objects include parts of the human body
• CPC Class Codes: H04N 23/651 for reducing power consumption by affecting camera operations, e.g. sleep mode, hibernation mode or power off of selective parts of the camera
• CPC Class Codes: H04N 23/6811 based on the image signal
• CPC Class Codes: H04N 23/74 by influencing the scene brightness using illuminating means
• CPC Class Codes: H04N 25/44 by partially reading an SSIS array

Abstract

Embodiments related to detecting object information from image data collected by an image sensor are disclosed. In one example embodiment, the object information is detected by receiving a frame of image data from the image sensor and detecting a change in a threshold condition related to an object within the frame. The embodiment further comprises adjusting a setting that changes a power consumption of the image sensor in response to detecting the threshold condition.

Claims

1. In a computing device, a method of detecting object information in image data collected by an image sensor, the method comprising:

receiving a frame of image data from the image sensor;

detecting a change in a threshold condition related to an object within the frame; and

adjusting a setting that changes a power consumption of the image sensor in response to detecting the change in the threshold condition.

2. The method of claim 1, wherein adjusting a setting comprises adjusting a hardware setting.

3. The method of claim 2, wherein adjusting the hardware setting comprises adjusting one or more of an illuminant setting, a resolution setting and a frame rate setting.

4. The method of claim 2, wherein adjusting the setting comprises increasing an amount of image data collected per unit time.

5. The method of claim 2, wherein adjusting the setting comprises decreasing an amount of image data collected per unit time.

6. The method of claim 2, wherein adjusting the setting comprises adjusting a software setting in an image processing pipeline, the software setting comprising one or more of a segmentation resolution setting, a classification resolution setting, and a model fitting resolution setting.

7. The method of claim 1, where detecting the change in the threshold condition comprises initially detecting the object within the frame.

8. The method of claim 1, further comprising:

recognizing the object after detecting the object;

identifying a region of the frame that does not include the object in response to recognizing the object; and

adjusting a setting that decreases an image data per unit time collected by the image sensor within the region that does not include the object in response to identifying the region.

9. The method of claim 8, where identifying the region includes identifying a candidate region that is projected to exclude the object in a subsequently-received frame based on object information extracted from image data included in a previously-received frame.

10. A computing system configured to receive image data from an image sensor and to extract object information from the image data, the computing system comprising:

a logic subsystem; and

a data holding subsystem storing instructions executable by the logic subsystem to;

receive from the image sensor a first image acquired at a first image collection setting,

recognize an object in the first image, and

in response to recognizing the object in the first image;

control the image sensor to change the first image data collection setting to a second image data collection setting in which more image data per unit time is collected than in the first image data collection setting, and

receive a second image from the image sensor acquired at the second image data collection setting.

11. The computing system of claim 10, further comprising instructions executable to adjust a setting in an image processing pipeline, the setting in the image processing pipeline comprising one or more of segmentation resolution information, classification resolution information, and model fitting resolution information.

12. The computing system of claim 10, where the instructions executable to adjust the image sensor setting include instructions that change power consumption for the image sensor.

13. The computing system of claim 12, where the instructions executable to change power consumption for the image sensor include instructions executable to increase power consumption for the image sensor.

14. The computing system of claim 12, where the instructions executable to change power consumption for the image sensor include instructions executable to change one or more of an illumination setting, a frame rate setting, and a resolution setting for the image sensor.

15. The computing system of claim 10, further comprising instructions executable to:

identify a background region of the first image that does not include the object in response to recognizing the object;

adjust a setting that decreases an image data per unit time collected by the image sensor within the background region in response to identifying the background region; and

identify a candidate region that is projected to exclude the object in a subsequently-collected image based on object information extracted from image data included in a previously-collected image.

16. A computer-readable storage medium comprising instructions stored thereon that are executable by a computing device to:

receive a first image;

recognize an object in the first image; and

responsive to recognizing the object in the first image;

adjust a value of an object information extraction parameter to an adjusted value with which more object information is extracted than with the first value, and

extract the object from a second image with the adjusted value.

17. The computer-readable storage medium of claim 16, further comprising instructions executable to:

wherein adjusting the object information extraction parameter comprises adjusting one or more of a segmentation resolution setting, a classification resolution setting, and a model fitting resolution setting.

18. The computer-readable storage medium of claim 16, further comprising instructions executable to adjust a setting for an image sensor that increases power consumption for the image sensor in response to recognizing the object in the first image.

19. The computer-readable storage medium of claim 16, further comprising instructions executable to control an image data collection setting for an image sensor to change a first image data collection setting to a second image data collection setting in which more image data per unit time is collected than in the first image data collection setting in response to recognizing the object in the first image.

20. The computer-readable storage medium of claim 19, further comprising instructions executable to:

identify a background region of the first image that does not include the object in response to recognizing the object;

adjust a setting that decreases an image data per unit time collected by the image sensor within the background region in response to identifying the background region; and

identify a candidate region that is projected to exclude the object in a subsequently-received image based on object information extracted from image data included in a previously-received image.