Abstract: Systems and methods for calibrating an array camera are disclosed. Systems and methods for calibrating an array camera in accordance with embodiments of this invention include the capturing of an image of a test pattern with the array camera such that each imaging component in the array camera captures an image of the test pattern. The image of the test pattern captured by a reference imaging component is then used to derive calibration information for the reference component. A corrected image of the test pattern for the reference component is then generated from the calibration information and the image of the test pattern captured by the reference imaging component. The corrected image is then used with the images captured by each of the associate imaging components associated with the reference component to generate calibration information for the associate imaging components.

Abstract: A hand-held or otherwise portable or spatial or temporal performance-based image capture device includes one or more lenses, an aperture and a main sensor for capturing an original main image. A secondary sensor and optical system are for capturing a reference image that has temporal and spatial overlap with the original image. The device performs an image processing method including capturing the main image with the main sensor and the reference image with the secondary sensor, and utilizing information from the reference image to enhance the main image. The main and secondary sensors are contained together within a housing.

Abstract: Systems and methods for implementing array cameras configured to perform super-resolution processing to generate higher resolution super-resolved images using a plurality of captured images and lens stack arrays that can be utilized in array cameras are disclosed. Lens stack arrays in accordance with many embodiments of the invention include lens elements formed on substrates separated by spacers, where the lens elements, substrates and spacers are configured to form a plurality of optical channels, at least one aperture located within each optical channel, at least one spectral filter located within each optical channel, where each spectral filter is configured to pass a specific spectral band of light, and light blocking materials located within the lens stack array to optically isolate the optical channels.

Abstract: A method of iris recognition comprises detecting a body region larger than and comprising at least one iris in an image and performing a first eye modelling on the detected body region. If successful, the result of first iris segmentation based on the first eye model is chosen. Otherwise, a first iris identification is performed on the detected body region. If successful, the result of second iris segmentation based on a second eye modelling is chosen. Otherwise, second iris identification is performed on the image, third eye modelling is performed on the result of the second iris identification, and third iris segmentation is performed on the result of the third eye modelling. If successful, the result of third iris segmentation based on a third eye modelling is chosen. An iris code is extracted from any selected iris segment of the image.

Abstract: A method for correcting an image divides an output image into a grid with vertical sections of width smaller than the image width but wide enough to allow efficient bursts when writing distortion corrected line sections into memory. A distortion correction engine includes a relatively small amount of memory for an input image buffer but without requiring unduly complex control. The input image buffer accommodates enough lines of an input image to cover the distortion of a single most vertically distorted line section of the input image. The memory required for the input image buffer can be significantly less than would be required to store all the lines of a distorted input image spanning a maximal distortion of a complete line within the input image.

Abstract: A method of tracking an object across a stream of images comprises determining a region of interest (ROI) bounding the object in an initial frame of an image stream. A HOG map is provided for the ROI by: dividing the ROI into an array of MxN cells, each cell comprising a plurality of image pixels; and determining a HOG for each of the cells. The HOG map is stored as indicative of the features of the object. Subsequent frames are acquired from the stream of images. The frames are scanned ROI by ROI to identify a candidate ROI having a HOG map best matching the stored HOG map features. If the match meets a threshold, the stored HOG map indicative of the features of the object is updated according to the HOG map for the best matching candidate ROI.

Abstract: A peripheral processing device comprises a physical interface for connecting the processing device to a host computing device through a communications protocol. A local controller connected to local memory across an internal bus provides input/output access to data stored on the processing device to the host through a file system API. A neural processor comprises at least one network processing engine for processing a layer of a neural network according to a network configuration. A memory at least temporarily stores network configuration information, input image information, intermediate image information and output information produced by each network processing engine. The local controller is arranged to receive network configuration information through a file system API write command, to receive input image information through a file system API write command; and to write output information to the local memory for retrieval by the host through a file system API read command.

Abstract: Tracking units for facial features with advanced training for natural rendering of human faces in real-time are provided. An example device receives a video stream, and upon detecting a visual face, selects a 3D model from a comprehensive set of head orientation classes. The device determines modifications to the selected 3D model to describe the face, then projects a 2D model of tracking points of facial features based on the 3D model, and controls, actuates, or animates hardware based on the facial features tracking points. The device can switch among an example comprehensive set of 35 different head orientation classes for each video frame, based on suggestions computed from a previous video frame or from yaw and pitch angles of the visual head orientation. Each class of the comprehensive set is trained separately based on a respective collection of automatically marked images for that head orientation class.

Abstract: Systems and methods for implementing array camera configurations that include a plurality of constituent array cameras, where each constituent array camera provides a distinct field of view and/or a distinct viewing direction, are described. In several embodiments, image data captured by the constituent array cameras is used to synthesize multiple images that are subsequently blended. In a number of embodiments, the blended images include a foveated region. In certain embodiments, the blended images possess a wider field of view than the fields of view of the multiple images.

Abstract: A method for automatically determining exposure settings for an image acquisition system comprises maintaining a plurality of look-up tables, each look-up table being associated with a corresponding light condition and storing image exposure settings associated with corresponding distance values between a subject and the image acquisition system. An image of a subject is acquired from a camera module; and a light condition occurring during the acquisition is determined based on the acquired image. A distance between the subject and the camera module during the acquisition is calculated.

Abstract: A method for compensating for off-axis tilting of a lens relative to an image sensor in an image acquisition device comprises acquiring a set of calibrated parameters n ? ? ( n x n y n z ) corresponding to the tilting of said lens. P?x and P?y indicate a coordinate of a pixel in an acquired image. Image information is mapped from the acquired image to a lens tilt compensated image according to the formulae: P x ? = s n z ? ( n z - 1 ) ? ( ( n x 2 + n z ? ( n z - 1 ) ) ? u x + n x ? n y ? u y ) P y ? = s n z ? ( n z - 1 ) ? ( ( n y 2 + n z ? ( n z - 1 ) ) ? u y + n x ? n y ? u x ) where s comprises a scale factor given by s = n z u x ? n x + u y ? n y + n z and where ux and uy indicate the location of a pixel in the lens tilt compensated image.

Abstract: Systems and methods for implementing array cameras configured to perform super-resolution processing to generate higher resolution super-resolved images using a plurality of captured images and lens stack arrays that can be utilized in array cameras are disclosed. An imaging device in accordance with one embodiment of the invention includes at least one imager array, and each imager in the array comprises a plurality of light sensing elements and a lens stack including at least one lens surface, where the lens stack is configured to form an image on the light sensing elements, control circuitry configured to capture images formed on the light sensing elements of each of the imagers, and a super-resolution processing module configured to generate at least one higher resolution super-resolved image using a plurality of the captured images.

Abstract: A neural network image processing apparatus arranged to acquire images from an image sensor and to: identify a ROI containing a face region in an image; determine at plurality of facial landmarks in the face region; use the facial landmarks to transform the face region within the ROI into a face region having a given pose; and use transformed landmarks within the transformed face region to identify a pair of eye regions within the transformed face region. Each identified eye region is fed to a respective first and second convolutional neural network, each network configured to produce a respective feature vector. Each feature vector is fed to respective eyelid opening level neural networks to obtain respective measures of eyelid opening for each eye region. The feature vectors are combined and to a gaze angle neural network to generate gaze yaw and pitch values substantially simultaneously with the eyelid opening values.

Abstract: A multi-processor neural network processing apparatus comprises: a plurality of network processing engines, each for processing one or more layers of a neural network according to a network configuration. A memory at least temporarily stores network configuration information, input image information, intermediate image information and output information for the network processing engines. At least one of the network processing engines is configured, when otherwise idle, to identify configuration information and input image information to be processed by another target network processing engine and to use the configuration information and input image information to replicate the processing of the target network processing engine.

Abstract: Systems with an array camera augmented with a conventional camera in accordance with embodiments of the invention are disclosed. In some embodiments, the array camera is used to capture a first set of image data of a scene and a conventional camera is used to capture a second set of image data for the scene. An object of interest is identified in the first set of image data. A first depth measurement for the object of interest is determined and compared to a predetermined threshold. If the first depth measurement is above the threshold, a second set of image data captured using the conventional camera is obtained. The object of interest is identified in the second set of image data and a second depth measurement for the object of interest is determined using at least a portion of the first set of image data and at least a portion of the second set of image data.

Abstract: In one embodiment, a gimbal adjustment system and an associated method for adjusting the position of an object. The system comprises a base, a plate and a shaft including a pivot attached to the plate. The pivot has a point of contact with the plate in a joint about which the plate is rotatable. Magnetic elements are positioned on the base and the plate to stabilize or rotate the plate. The object may be an optical unit attached to the plate. A combination comprising the plate, optical unit and magnetic elements may form a gimbaled assembly having a center of mass in the joint.

Abstract: A method of providing a sharpness measure for an image comprises detecting an object region within an image; obtaining meta-data for the image; and scaling the chosen object region to a fixed size. A gradient map is calculated for the scaled object region and compared against a threshold determined for the image to provide a filtered gradient map of values exceeding the threshold. The threshold for the image is a function of at least: a contrast level for the detected object region, a distance to the subject and an ISO/gain used for image acquisition. A sharpness measure for the object region is determined as a function of the filtered gradient map values, the sharpness measure being proportional to the filtered gradient map values.

Abstract: An optical system for a shallow form factor camera module comprises: seven even aspheric singlet lens elements formed of a molded plastics material, a flat image sensor, and a central aperture stop with at least 3 lens elements disposed in front of the aperture stop and at least 3 rear lens elements disposed between the aperture stop and the image sensor. The aperture stop has a pupil wide enough to provide at least f/1.0 aperture and the optical system has a total track length less than about 8.2 mm and a minimum spacing of about 0.6 mm between a surface of the rearmost lens element and an imaging plane of said image sensor. A magnitude of an aspheric coefficient for h4 of at least one surface of each of said rear lens elements is greater than 5×10?2.

Abstract: An image processing apparatus comprises a set of infra-red (IR) sources surrounding an image capture sensor and a processor operatively coupled to said IR sources and said image capture sensor. The processor being arranged to acquire from the sensor a succession of images, each illuminated with a different combination of the IR sources. The processor is further arranged to combine component images corresponding to the succession of images by selecting a median value for corresponding pixel locations of the component images as a pixel value for the combined image.

Abstract: Systems and methods for hybrid depth regularization in accordance with various embodiments of the invention are disclosed. In one embodiment of the invention, a depth sensing system comprises a plurality of cameras; a processor; and a memory containing an image processing application. The image processing application may direct the processor to obtain image data for a plurality of images from multiple viewpoints, the image data comprising a reference image and at least one alternate view image; generate a raw depth map using a first depth estimation process, and a confidence map; and generate a regularized depth map. The regularized depth map may be generated by computing a secondary depth map using a second different depth estimation process; and computing a composite depth map by selecting depth estimates from the raw depth map and the secondary depth map based on the confidence map.