Methods and apparatus to operate a mobile camera for low-power usage

Abstract

Disclosed examples include accessing sensor data; recognizing, by executing an instruction with programmable circuitry, a feature in the sensor data based on a convolutional neural network; and transitioning, by executing an instruction with the programmable circuitry, a mobile device between at least two of motion feature detection, audio feature detection, or camera feature detection after the feature is recognized in the sensor data, the mobile device to operate at a different level of power consumption after the transition than before the transition.

What is claimed is: 
     
         1 . A mobile device comprising:
 a camera;   an interface to receive sensor data;   a convolutional neural network to recognize a feature in the sensor data;   programmable circuitry; and   instructions to cause the programmable circuitry to transition the mobile device between at least two of motion feature detection, audio feature detection, or camera feature detection after the convolutional neural network recognizes the feature in the sensor data, the mobile device to operate at a different level of power consumption after the transition than before the transition.   
     
     
         2 . The mobile device of  claim 1 , further including a communications interface to send metadata to an external device, the metadata corresponding to the feature. 
     
     
         3 . The mobile device of  claim 2 , wherein the convolutional neural network is to generate the metadata by comparing the feature to a reference feature definition in a reference metadata library. 
     
     
         4 . The mobile device of  claim 1 , wherein the camera feature detection corresponds to low-resolution feature detection. 
     
     
         5 . The mobile device of  claim 1 , wherein the camera feature detection corresponds to a still image. 
     
     
         6 . The mobile device of  claim 1 , wherein the programmable circuitry is to perform the transition of the mobile device after the feature satisfies a feature trigger threshold. 
     
     
         7 . The mobile device of  claim 1 , wherein the feature corresponds to at least one of: (a) speech, (b) a vehicle sound, or (c) a rate of change in an audio signal. 
     
     
         8 . A non-transitory computer readable storage medium comprising instructions to cause programmable circuitry to at least:
 access sensor data;   execute a convolutional neural network to recognize a feature in the sensor data; and   transition a mobile device between at least two of motion feature detection, audio feature detection, or camera feature detection after the convolutional neural network recognizes the feature in the sensor data, the mobile device to operate at a different level of power consumption after the transition than before the transition.   
     
     
         9 . The non-transitory computer readable storage medium of  claim 8 , wherein the instructions are to cause the programmable circuitry to send metadata to an external device, the metadata corresponding to the feature. 
     
     
         10 . The non-transitory computer readable storage medium of  claim 9 , wherein the instructions are to cause the programmable circuitry to execute the convolutional neural network to generate the metadata by comparing the feature to a reference feature definition in a reference metadata library. 
     
     
         11 . The non-transitory computer readable storage medium of  claim 8 , wherein the camera feature detection corresponds to low-resolution feature detection. 
     
     
         12 . The non-transitory computer readable storage medium of  claim 8 , wherein the camera feature detection corresponds to a still image. 
     
     
         13 . The non-transitory computer readable storage medium of  claim 8 , wherein the instructions are to cause the programmable circuitry to perform the transition of the mobile device after the feature satisfies a feature trigger threshold. 
     
     
         14 . The non-transitory computer readable storage medium of  claim 8 , wherein the feature corresponds to at least one of: (a) speech, (b) a vehicle sound, or (c) a rate of change in an audio signal. 
     
     
         15 . A method comprising:
 accessing sensor data;   recognizing, by executing an instruction with programmable circuitry, a feature in the sensor data based on a convolutional neural network; and   transitioning, by executing an instruction with the programmable circuitry, a mobile device between at least two of motion feature detection, audio feature detection, or camera feature detection after the feature is recognized in the sensor data, the mobile device to operate at a different level of power consumption after the transition than before the transition.   
     
     
         16 . The method of  claim 15 , further including transmitting metadata to an external device, the metadata corresponding to the feature. 
     
     
         17 . The method of  claim 16 , further including generating the metadata by comparing the feature to a reference feature definition in a reference metadata library. 
     
     
         18 . The method of  claim 15 , wherein the camera feature detection corresponds to low-resolution feature detection. 
     
     
         19 . The method of  claim 15 , wherein the transitioning of the mobile device is after the feature satisfies a feature trigger threshold. 
     
     
         20 . The method of  claim 15 , wherein the feature corresponds to at least one of: (a) speech, (b) a vehicle sound, or (c) a rate of change in an audio signal.