Provided is an apparatus for learning human pose estimation by configuring a dataset for human pose estimation by simultaneously obtaining a well-lit image and a low-light image, performing annotation in the well-lit image, and transmitting the annotation to the low-light image. By using the well-lit image included in the dataset as an input of a teacher model and the low-light image as an input of a student model, the student model learns human pose estimation at a high accuracy in low-light conditions by using privileged information of the teacher model.

Imager arrays, array camera modules, and array cameras in accordance with embodiments of the invention utilize pixel apertures to control the amount of aliasing present in captured images of a scene. One embodiment includes a plurality of focal planes, control circuitry configured to control the capture of image information by the pixels within the focal planes, and sampling circuitry configured to convert pixel outputs into digital pixel data. In addition, the pixels in the plurality of focal planes include a pixel stack including a microlens and an active area, where light incident on the surface of the microlens is focused onto the active area by the microlens and the active area samples the incident light to capture image information, and the pixel stack defines a pixel area and includes a pixel aperture, where the size of the pixel apertures is smaller than the pixel area.

Snapshot spectral imagers comprise an imaging lens, a dispersed image sensor and a restricted isometry property (RIP) diffuser inserted in the optical path between the source image and the image sensor. The imagers are used to obtain a plurality of spectral images of the source object in different spectral bands in a single shot. In some embodiments, the RIP diffuser is one dimensional. An optional disperser may be added in the optical path, to provide further dispersion at the image sensor. In some embodiments, all imager components except the RIP diffuser may be part of a digital camera, with the RIP diffuser added externally. In some embodiments, the RIP diffuser may be included internally in a digital camera.

Spatio-temporal light field cameras that can be used to capture the light field within its spatio temporally extended angular extent. Such cameras can be used to record 3D images, 2D images that can be computationally focused, or wide angle panoramic 2D images with relatively high spatial and directional resolutions. The light field cameras can be also be used as 2D/3D switchable cameras with extended angular extent. The spatio-temporal aspects of the novel light field cameras allow them to capture and digitally record the intensity and color from multiple directional views within a wide angle. The inherent volumetric compactness of the light field cameras make it possible to embed in small mobile devices to capture either 3D images or computationally focusable 2D images. The inherent versatility of these light field cameras makes them suitable for multiple perspective light field capture for 3D movies and video recording applications.

Array cameras, and array camera modules incorporating independently aligned lens stacks are disclosed. Processes for manufacturing array camera modules including independently aligned lens stacks can include: forming at least one hole in at least one carrier; mounting the at least one carrier relative to at least one sensor so that light passing through the at least one hole in the at least one carrier is incident on a plurality of focal planes formed by arrays of pixels on the at least one sensor; and independently mounting a plurality of lens barrels to the at least one carrier, so that a lens stack in each lens barrel directs light through the at least one hole in the at least one carrier and focuses the light onto one of the plurality of focal planes.

A system, method, and computer program product for generating a digital image is disclosed. The system includes a first image sensor configured to capture a first image that includes a plurality of chrominance values, a second image sensor configured to capture a second image that includes a plurality of luminance values, and an image processing subsystem configured to generate a resulting image by combining the plurality of chrominance values with the plurality of luminance values. The first image sensor and the second image sensor may be distinct image sensors optimized for capturing chrominance images or luminance images.

Provided is a parallax minimization stitching method and apparatus using control points in an overlapping region. A parallax minimization stitching method may include defining a plurality of control points in an overlapping region of a first image and a second image received from a plurality of cameras, performing a first geometric correction by applying a homography to the control points, defining a plurality of patches based on the control points, and performing a second geometric correction by mapping the patches.

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.

A multi-aperture imaging system comprising a first camera with a first sensor that captures a first image and a second camera with a second sensor that captures a second image, the two cameras having either identical or different FOVs. The first sensor may have a standard color filter array (CFA) covering one sensor section and a non-standard color CFA covering another. The second sensor may have either Clear or standard CFA covered sections. Either image may be chosen to be a primary or an auxiliary image, based on a zoom factor. An output image with a point of view determined by the primary image is obtained by registering the auxiliary image to the primary image.

An image acquisition apparatus may include a multispectral image sensor configured to acquire an image of at least one object in an environment in which at least one illumination source exists, through eight or more channels with minimum overlap between the channels, and a processor configured to estimate illumination spectral data of the acquired image by using channel signals corresponding to the eight or more channels, and perform lens shading correction on the acquired image, based on the estimated illumination spectral data.