Back to results
US 11,240,430 B2Granted
Methods and apparatus to operate a mobile camera for low-power usage
AssigneeMovidius Limited
InventorsDavid Moloney, Alireza Dehghani
Life of this patent
Estimating the term…
Abstract
A disclosed example to operate a mobile camera includes recognizing a first feature in first sensor data in response to the first feature being detected in the first sensor data; transitioning the mobile camera from a first feature detection state to a second feature detection state in response to the recognizing of the first feature, the mobile camera to operate using higher power consumption in second feature detection state than in the first feature detection state; recognizing a second feature in second sensor data in the second feature detection state; and sending to an external device at least one of first metadata corresponding to the first feature or second metadata corresponding to the second feature.
Claims25
A mobile camera, comprising:
- a first state transitioner to transition the mobile camera from a motion feature detection state to an audio feature detection state in response to a motion feature in motion sensor data, a first power state of the motion feature detection state corresponding to lower power consumption than a second power state of the audio feature detection state;
- a second state transitioner to transition the mobile camera from the audio feature detection state to a low-resolution camera feature detection state in response to a first convolutional neural network recognizing an audio feature in audio data, the mobile camera to operate using higher power consumption in a third power state corresponding to the low-resolution camera feature detection state than in the second power state corresponding to the audio feature detection state;
- and
- a communications interface to send to an external device at least one of:
- (a) first metadata corresponding to the motion feature or the audio feature, or (b) second metadata corresponding to an image feature recognized in visual capture data by a second convolutional neural network when the mobile camera is in the low-resolution camera feature detection state.
2 more elements
An apparatus, comprising:
- first circuitry to transition a mobile camera from a motion feature detection state to an audio feature detection state in response to a motion feature in motion sensor data, a first power state of the motion feature detection state corresponding to lower power consumption than a second power state of the audio feature detection state;
- second circuitry to transition the mobile camera from the audio feature detection state to a first camera feature detection state in response to a first convolutional neural network recognizing an audio feature in audio data, the mobile camera to operate using higher power consumption in a third power state corresponding to the first camera feature detection state than in the second power state corresponding to the audio feature detection state;
- and
- communications circuitry to transmit at least one of:
- (a) first metadata corresponding to the motion feature or the audio feature, or (b) second metadata corresponding to an image feature recognized in visual capture data by a second convolutional neural network when the mobile camera is in the first camera feature detection state.
2 more elements
A non-transitory computer readable storage medium comprising instructions that, when executed, cause at least one processor to at least:
- transition a mobile camera from a motion feature detection state to an audio feature detection state in response to a motion feature in motion sensor data, a first power state of the motion feature detection state corresponding to lower power consumption than a second power state of the audio feature detection state;
- transition the mobile camera from the audio feature detection state to a low-resolution camera feature detection state in response to a first convolutional neural network recognizing an audio feature in audio data, the mobile camera to operate using higher power consumption in a third power state corresponding to the low-resolution camera feature detection state than in the second power state corresponding to the audio feature detection state;
- and
- send to an external device at least one of:
- (a) first metadata corresponding to the motion feature or the audio feature, or (b) second metadata corresponding to an image feature recognized in visual capture data by a second convolutional neural network when the mobile camera is in the low-resolution camera feature detection state.
2 more elements
A method to operate a mobile camera, the method comprising:
- transitioning the mobile camera from a motion feature detection state to an audio feature detection state in response to a motion feature in motion sensor data, a first power state of the motion feature detection state corresponding to lower power consumption than a second power state of the audio feature detection state;
- transitioning the mobile camera from the audio feature detection state to a low-resolution camera feature detection state in response to a first convolutional neural network recognizing an audio feature in audio data, the mobile camera to operate using higher power consumption in a third power state corresponding to the low-resolution camera feature detection state than in the second power state corresponding to the audio feature detection state;
- and
- sending to an external device at least one of:
- (a) first metadata corresponding to the motion feature or the audio feature, or (b) second metadata corresponding to an image feature recognized in visual capture data by a second convolutional neural network when the mobile camera is in the low-resolution camera feature detection state.
2 more elements
Dependencies are inferred from the claim text.