Wearable information system having at least one camera

Abstract

The invention is related to a wearable information system having at least one camera, the information system operable to have a low-power mode and a high power mode. The information system is configured such that the high-power mode is activated by a detection of at least one object in at least one field of view of the at least one camera.

What is claimed is: 
     
         1 .- 21 . (canceled) 
     
     
         22 . An information system, comprising:
 a camera;   a processor operatively coupled to the camera;   a device operatively coupled to the processor; and   a memory device operatively coupled to the camera, the processor and the device, the memory device comprising instructions executable by the processor to:
 obtain, in a low-power mode of the information system, an image captured by the camera; 
 extract, in the low-power mode, a first feature of an object in the image; 
 generate, in the low-power mode, a higher level descriptor of the first feature; 
 cause, in the low-power mode, the device to determine that the higher level descriptor matches a reference object feature descriptor; and 
 activate, in response to determining the higher level descriptor matches the reference object feature descriptor, a high-power mode of the information system. 
   
     
     
         23 . The information system of  claim 22 , wherein the device determines that at least one of the higher level descriptors matches a reference object feature descriptor by:
 loading a plurality of reference object feature descriptors into a memory of the device;   loading the higher level descriptor for the first feature;   determining a distance measure between the higher level descriptor and each of the plurality of reference object feature descriptors; and   calculating a check parameter to determine whether the higher level descriptor is a valid match for at least one of the plurality of reference object feature descriptors.   
     
     
         24 . The information system of  claim 22 , wherein in the low-power mode a clock rate of the processor is lower than in the high-power mode. 
     
     
         25 . The information system of  claim 22 , wherein the first feature comprises a point-feature. 
     
     
         26 . The information system according to  claim 22 , wherein the higher level descriptor comprises a scale-invariant feature descriptor. 
     
     
         27 . The information system according to  claim 22 , wherein the higher level descriptor comprises a rotation-invariant feature descriptor. 
     
     
         28 . The information system according to  claim 22 , further comprising instructions to cause the processor to display, in the high-power mode, augmented reality information related to the object. 
     
     
         29 . The information system of  claim 22 , wherein the processor comprises one or more processors. 
     
     
         30 . The information system of  claim 29 , wherein the instructions to cause the one or more processors to extract a first feature of an object in the image comprise instructions to cause the one or more processors to extract one or more features of the object in the image. 
     
     
         31 . A computer readable medium comprising computer readable code executable by a processor to:
 obtain, in a low-power mode of a system, an image captured by the camera;   extract, in the low-power mode, a first feature of an object in the image;   generate, in the low-power mode, a higher level descriptor of the first feature;   cause, in the low-power mode, a device to determine that the higher level descriptor matches a reference object feature descriptor; and   activate, in response to determining the higher level descriptor matches the reference object feature descriptor, a high-power mode of the information system.   
     
     
         32 . The computer readable medium of  claim 31 , wherein the device determines that at least one of the higher level descriptors matches a reference object feature descriptor by:
 loading a plurality of reference object feature descriptors into a memory of the device;   loading the higher level descriptor for the first feature;   determining a distance measure between the higher level descriptor and each of the plurality of reference object feature descriptors; and   calculating a check parameter to determine whether the higher level descriptor is a valid match for at least one of the plurality of reference object feature descriptors.   
     
     
         33 . The computer readable medium of  claim 31 , wherein in the low-power mode a clock rate of the processor is lower than in the high-power mode. 
     
     
         34 . The computer readable medium of  claim 31 , wherein the first feature comprises a point-feature. 
     
     
         35 . The computer readable medium of  claim 31 , wherein the higher level descriptor comprises a scale-invariant feature descriptor. 
     
     
         36 . The computer readable medium of  claim 31 , wherein the higher level descriptor comprises a rotation-invariant feature descriptor. 
     
     
         37 . The computer readable medium of  claim 31 , further comprising computer readable code to cause the processor to display, in the high-power mode, augmented reality information related to the object. 
     
     
         38 . A method for managing a low-power and high-power mode of a system, comprising:
 obtaining, in a low-power mode of a system, an image captured by the camera;   extracting, in the low-power mode, a first feature of an object in the image;   generating, in the low-power mode, a higher level descriptor of the first feature;   causing, in the low-power mode, a device to determine that the higher level descriptor matches a reference object feature descriptor; and   activating, in response to determining the higher level descriptor matches the reference object feature descriptor, a high-power mode of the information system.   
     
     
         39 . The method of  claim 38 , wherein the device determines that at least one of the higher level descriptors matches a reference object feature descriptor by:
 loading a plurality of reference object feature descriptors into a memory of the device;   loading the higher level descriptor for the first feature;   determining a distance measure between the higher level descriptor and each of the plurality of reference object feature descriptors; and   calculating a check parameter to determine whether the higher level descriptor is a valid match for at least one of the plurality of reference object feature descriptors.   
     
     
         40 . The method of  claim 38 , wherein the higher level descriptor comprises a scale-invariant feature descriptor. 
     
     
         41 . The method of  claim 38 , further comprising displaying, in the high-power mode, augmented reality information related to the object.