Abstract: Embodiments enable detecting an occurrence such as an impact event, which can occur when an electronic device impacts a surface or object (e.g., from a fall), which may have occurred while the device was powered off, in a low power mode, or other state where the device cannot detect such an impact using an accelerometer or other such sensor. Various calibration processes can be used to determine an amount of misalignment between two or more cameras of the device, where an amount of misalignment more than an allowable threshold can be indicative of the impact event, at which point one or more system checks can be performed to determine whether the device components (e.g., memory, hard-disk, drives, antennas, etc.) or certain portions or components of the device are operating properly.

Abstract: Approaches are described which enable a computing device (e.g., mobile phone, tablet computer) to utilize one or more facial recognition techniques to control access to the device and to detect when artificial representations of a user, such as a picture or photograph, are being used in an attempt to gain access to the device. Evidence indicative of artificial representations may include lack of changes in facial skin color between multiple images captured by a camera, ability to track one or more features of the human face while the camera is rotated or moved, presence of secular reflections caused by an illumination device, absence of shadows in the image, and others.

Abstract: The location of a user's head, for purposes such as head tracking or motion input, can be determined using a two-step process. In a first step, at least one image is captured including a representation of at least a portion of the person, such as a head portion of the person. In a second step, a contour of the head portion can be determined, and a two-dimensional model, for example, an ellipse or other similar shape can be used to approximate the head portion of the person represented in the image. The ellipse, for example, can be modeled using a number of shapes, such as rectangles, and the portion of the person can be tracked by locating an ellipse that bounds a maximum intensity gradient of pixel values in each one of a series of images.

Abstract: The amount of power and processing needed to enable more previze gesture input for a computing device can be reduced by utilizing one or more gesture sensors along with a conventional camera. Information associated with an image, such as color information, acquired from the conventional camera can be mapped to one or more gesture sensors (e.g., low resolution, greyscale or monochrome camera sensors) of a computing device. The mapped information between the conventional camera and the gesture sensors can be used for purposes such as gesture- and/or motion-based approaches.

Abstract: Various embodiments enable a primary user to be identified and tracked using stereo association and multiple tracking algorithms. For example, a face detection algorithm can be run on each image captured by a respective camera independently. Stereo association can be performed to match faces between cameras. If the faces are matched and a primary user is determined, a face pair is created and used as the first data point in memory for initializing object tracking. Further, features of a user's face can be extracted and the change in position of these features between images can determine what tracking method will be used for that particular frame.

Abstract: A computing device with multiple image capture elements can selectively activate those elements in order to keep an object of interest within the field of view of at least one of those elements, while conserving resources by not keeping all the image capture elements active. The object can be located using an appropriate object recognition process. The location, and thus the motion, of the object then can be tracked over time using one or the image capture elements. The motion can be monitored to determine when the object is likely to pass into the field of view of another image capture element. The determination can be made based upon factors such as location, direction, speed, acceleration, and predicted time within the current field of view. Such an approach allows an identified object to be tracked and kept in a field of view while conserving resources on the device.

Abstract: Depth information can be used to assist with image processing functionality, such as various calibration approaches to determine and correct for misalignments between cameras. In at least some embodiments, depth information obtained from stereo imaging or distance sensing, for example, can be used to segment an image or frame of video into at least a foreground object and a background object. Once the foreground and background object has been determined, information about that objects can be used to determine a mapping, and once a subsequent stereoscopic image is captured using the cameras, the mapping can be applied to each image to account for misalignment effects due to a misaligned stereo camera pair before providing the stereoscopic image for display.

Abstract: The quality of stereoscopic imaging using cameras with wide angle lenses can be improved using various calibration approaches to determine distortions or misalignments between the cameras. A calibration object such as a checkered grid with a determined inclination between portions can be used to provide lateral calibration as well as depth information. Intersections of the checkered regions can be used to determine intersection points for the calibration object in the image. Regions about at least a portion of these points can be analyzed, where the regions are sized such that features of the calibration object are substantially linear. Nearest neighbor points can be identified, and a local orientation determined in order to determine a relationship of each of the nearest points. Once identified, the points in each image can be correlated to corresponding points of the calibration object.

Abstract: The quality of stereoscopic imaging using cameras with wide angle lenses can be improved using various calibration approaches to determine distortions or misalignments between the cameras. A calibration object such as a checkered grid with a determined inclination between portions can be used to provide lateral calibration as well as depth information. Intersections of the checkered regions can be used to determine intersection points for the calibration object in the image. Lines formed by the checkered regions can be located, and the points correlated with these lines, in order to determine correspondence between points in the image. These points are mapped to corresponding points of the calibration object in the real world, to determine system parameters and/or image adjustments to be made when subsequent images are captured, in order to remove the distortions and misalignment when providing a stereo image based on images from the cameras.

Abstract: The accuracy of object tracking using relatively low power gesture cameras can be improved by adjusting camera settings to ensure a sufficient level of contrast or texture to enable stereo disparity calculations for the representations of the object in the captured images. For general gesture input, for example, a user might use a variety of objects in various orientations, such that conventional face or object recognition processes may not be sufficient. Further, such processes are very resource intensive. By adjusting the camera settings (e.g., exposure, gain, and/or aperture) to ensure an adequate level of contrast, objects of an appropriate size and location can be tracked for purposes such as gesture input. Once such an object is identified, coordinates for a bounding box or other indicator can be transferred to any camera or sensor to be used for the tracking.

Abstract: Various embodiments utilize two-dimensional (“2D”) and three-dimensional (“3D”) object features for purposes such as object recognition and/or image matching. For example, a user can capture an image (e.g., still images or video) of an object and can receive information about items that are determined to match the object. For example, the image can be analyzed to detect visual features (e.g., corners, edges, etc.) of the object and the detected visual features can be combined to generate a combined visual feature vector which can be used for object recognition, image matching, or other such purposes. Other approaches utilize the image to generate a 3D model of the object represented in the image, which can be used to determine at least one object or types of objects that match the object represented in the image.

Abstract: Systems and approaches are provided for robustly determining the motion of a computing device. Multiple cameras on the device can each capture a sequence of images, and the images can be analyzed to determine motion of the device with respect to a user, an object, or scenery captured in the images. The estimated motion may be complemented with measurements from an inertial sensor such as a gyroscope or an accelerometer to provide more accurate estimations of device motion than can be provided by image data or inertial sensor data alone. A computing device can then be configured to detect device motion as user input such as to navigate a user interface or to remotely control movement of another electronic device.

Abstract: An electronic device can have two or more pairs of cameras capable of performing three-dimensional imaging. In order to provide accurate disparity information, these cameras should be sufficiently rectified. Automatic rectification can be performed by periodically capturing images with the cameras of interest, and locating matching feature points in corresponding images captured by those cameras. Small misalignment errors can be treated as linear translations, such that a set of linear equations can be used to solve for the misalignments. Another process can process a set of homographies for the cameras until a cost function converges. Various other approaches can be used as well, such as to directly solve for yaw, pitch, and roll errors. Once this information is obtained, the misalignment values (or related values) can be stored for use in correcting images subsequently captured by those cameras.

Abstract: Utilizing viewpoint and scale invariant feature descriptors for object recognition and/or tracking. For example, a user can capture three-dimensional (“3D”) image data including a representation of an object. The 3D image data can be analyzed to detect one or more feature points of the object represented in the image data, where the detected feature points can include position and distance information. The feature points in each image can be correlated and a feature descriptor, or unique fingerprint, can be determined for each detected feature point. The feature descriptors can provide for a multi-dimension vector that includes a unique fingerprint for that feature point that incorporates position information as well as depth information. The feature descriptors corresponding to the correlated feature points can be combined (e.g., added, averaged, etc.) and the combined feature descriptors can be used for performing view point invariant object recognition, image matching, or other such purposes.

Abstract: Approaches to enable a computing device, such as a phone or tablet computer, to determine depth information about an object captured by a single camera of the device without requiring multiple cameras to capture that object in their field of view. For example, the computing device may track the object throughout a sequence of images using a single rear-facing camera and then determine the depth information of that object by combining data about the object's changes in size (throughout the sequence of images) and information about the distance between the computing device and a user's face (or other user's feature) that can be determined using two or more front-facing cameras configured for stereo imaging. The depth information of the object may include the distance between the computing device and the object, the distance between the object and the user's face, or the physical dimensions of the object, among other such information.

Abstract: Various embodiments enable a primary user to be identified and tracked using stereo association and multiple tracking algorithms. For example, a face detection algorithm can be run on each image captured by a respective camera independently. Stereo association can be performed to match faces between cameras. If the faces are matched and a primary user is determined, a face pair is created and used as the first data point in memory for initializing object tracking. Further, features of a user's face can be extracted and the change in position of these features between images can determine what tracking method will be used for that particular frame.

Abstract: Approaches are described which enable a computing device (e.g., mobile phone, tablet computer) to utilize one or more facial recognition techniques to control access to the device and to detect when artificial representations of a user, such as a picture or photograph, are being used in an attempt to gain access to the device. Evidence indicative of artificial representations may include lack of changes in facial skin color between multiple images captured by a camera, ability to track one or more features of the human face while the camera is rotated or moved, presence of secular reflections caused by an illumination device, absence of shadows in the image, and others.

Abstract: Systems and approaches are provided for tracking an object of interest using depth or disparity information, such as obtained by calculating stereo disparity between a pair of images. The depth or disparity information can be used as an additional signature for a template of the object of interest for tracking the object. A template that includes depth, distance, or disparity information for an object of interest may be invariant to the effects of lighting, such as shadows and changes in illumination conditions. Depth, distance, or disparity information can also provide information regarding shape and size that can be used to differentiate foreground objects. Depth, distance, or disparity information can also better handle occlusion. Depth, distance, or disparity information can also provide an additional disambiguating dimension for tracking an object.

Abstract: A method for assisting in the recovery of a lost electronic device is described. When an electronic device determines that it is at rest, it automatically captures an image using a built-in camera and transmits the image to a server. A user, having lost the electronic device, may access the image via a web interface to determine the location of the electronic device.

Abstract: Systems and methods for implementing location-based security protocols for information and/or data files are disclosed. The location-based security protocols may be enforced by an organization to protect and/or provide additional levels of security for information and/or data files associated with the organization and stored and/or used by agents of the organization on their user devices, such as personal user devices. Location-based security protocols may be applied to data file functions that include generating, receiving, transmitting, sharing, backing-up, or rendering a data file. Data files that are to adhere to one or more location-based security protocols may be tagged, such as in metadata associated with the data file to indicate that it is subject to adherence to the one or more location-based security protocols.