Abstract:
System and methods for performing context inference in a computing device are disclosed. In one embodiment, a method of performing context inference includes: determining, at a computing device, a first context class using context-related data from at least one data source associated with a mobile device; and determining, at the mobile device, a fusion class based on the first context class, the fusion class being associated with at least one characteristic that is common to the first context class and a second context class that is different from the first context class.
Abstract:
Disclosed is an apparatus and method for power efficient processor scheduling of features. In one embodiment, features may be scheduled for sequential computing, and each scheduled feature may receive a sensor data sample as input. In one embodiment, scheduling may be based at least in part on each respective feature's estimated power usage. In one embodiment, a first feature in the sequential schedule of features may be computed and before computing a second feature in the sequential schedule of features, a termination condition may be evaluated.
Abstract:
Methods, systems, computer-readable media, and apparatuses for inferring context are provided. In one potential implementation, first context information associated with a first duration is identified, second context information is accessed to determine a context segmentation boundary; and the first context information and the second context information is then aggregated to generate an inferred segmented aggregated context. In a further implementation, the first context information is used to average inferred contexts, and the context segmentation boundary is used to reset a start time for averaging the first context information.
Abstract:
A context aware system, for use in a mobile device, includes a context change detector (CCD) coupled to a context classifier (CCL). The CCD is configured to receive sensor data and to detect a change in a current context state of the mobile device based on the received sensor data. The CCL is configured to transition from a low power consumption mode to a normal power consumption mode in response to the CCD detecting the change in the current context state. The CCL is further configured to determine a next context state of the mobile device while in the normal power consumption mode.
Abstract:
Systems and methods for applying and using context labels for data clusters are provided herein. A method described herein for managing a context model associated with a mobile device includes obtaining first data points associated with a first data stream assigned to one or more first data sources; assigning ones of the first data points to respective clusters of a set of clusters such that each cluster is respectively assigned ones of the first data points that exhibit a threshold amount of similarity and are associated with times within a threshold amount of time of each other; compiling statistical features and inferences corresponding to the first data stream or one or more other data streams assigned to respective other data sources; assigning context labels to each of the set of clusters based on the statistical features and inferences.
Abstract:
Sensor data from a sensor system of a mobile device may be used for determining a level of pressure exerted by a user on the mobile device. The sensor system may include one or more types of sensors, such as a microphone and one or more inertial sensors. The inertial sensors may include one or more gyroscopes and/or accelerometers. Based on the inertial sensor data, it may be determined whether and/or how the mobile device is being held. A process for determining a level of pressure exerted by a user on the mobile device may be adapted based, at least in part, on whether and/or how the mobile device is being held. The pressure-determining process may be adapted according to various other criteria, such as a position of a touch target in a display, ambient noise levels, etc.
Abstract:
Disclosed is an apparatus and method for classifying a motion state of a mobile device comprising: determining a first motion state associated with a highest probability value and with a first confidence level greater than a first threshold; entering the first motion state; while the first motion state is active, determining a second motion state associated with a highest probability value and with a second confidence level greater than the first threshold, the second motion state being different from the first motion state; determining whether the second motion state is to be entered; and in response to determining that the second motion state is to be entered, entering the second motion state.
Abstract:
Systems, apparatus and methods in a mobile device to enable and disable a depth sensor for tracking pose of the mobile device are presented. A mobile device relaying on a camera without a depth sensor may provide inadequate pose estimates, for example, in low light situations. A mobile device with a depth sensor uses substantial power when the depth sensor is enabled. Embodiments described herein enable a depth sensor only when images are expected to be inadequate, for example, accelerating or moving too fast, when inertial sensor measurements are too noisy, light levels are too low or high, an image is too blurry, or a rate of images is too slow. By only using a depth sensor when images are expected to be inadequate, battery power in the mobile device may be conserved and pose estimations may still be maintained.
Abstract:
Embodiments of the present invention are directed toward providing intelligent sampling strategies that make efficient use of an always-on camera. To do so, embodiments can utilize sensor information to determine contextual information regarding the mobile device and/or a user of the mobile device. A sampling rate of the always-on camera can then be modulated based on the contextual information.