Abstract: Methods, systems, and media for swapping faces in images are provided. In some embodiments, a detected face and face data corresponding to an input image is received. A pose bin associated with the detected face is then identified based on the face data. Next, the detected face is aligned to a generic face associated with the pose bin. At least a portion of a candidate face associated with the pose bin is selected. The at least a portion of the candidate face is then copied to a copy of the input image that is aligned with the generic image to form a swapped-face image. The swapped-face image is next aligned to the input image to form an output image, and then the output image is outputted to a display.

Abstract: The application provides techniques for obtaining a relatively high dynamic range image of a scene using a relatively low dynamic range image sensor exposed to incident light from the scene for capturing an image. The image sensor has a multiplicity of light-sensing elements in an array and each light sensing element has a particular one of a plurality of sensitivity levels to incident light in accordance with a predetermined sensitivity pattern for the array of light-sensing elements and has a response function. Each light sensing element is responsive to incident light from the scene for producing a captured image brightness value at a corresponding one of a multiplicity of pixel positions of a pixel position array. Each one of the multiplicity of pixel positions corresponds to a particular one of the plurality of sensitivity levels of the light sensing elements.

Abstract: Methods, systems, and media for automatically classifying face images are provided. In some embodiments, features of the face image to be classified for an attribute are selected, wherein each of the features corresponds to a different region of the face image and specifies one or more of a type of pixel data to be evaluated for the region, a normalization to be applied for the region, and an aggregation to be applied for the region. The face image is classified with respect to the attribute based on the features of the image, and the attribute and a confidence value are assigned to the face image based on the classifying. A query is received from a user, and the attribute is identified as corresponding to the query. The face image is determined as corresponding to the attribute, and the face image is identified to the user as corresponding to the query.

Abstract: Systems and methods for providing a substantially de-blurred image of a scene from a motion blurred image of the scene are disclosed. An exemplary system includes a primary detector for sensing the motion blurred image and generating primary image information representing the blurred image, a secondary detector for sensing two or more secondary images of the scene and for generating secondary image information representing the two or more secondary images, and a processor for determining motion information from the secondary image information, estimating a point spread function for the motion blurred image from the motion information, and applying the estimated point spread function to the primary image information to generate information representing the substantially de-blurred image.

Abstract: Systems, methods, and media for recording an image of a scene ate provided. In accordance with some embodiments, systems for recording an image of a scene are provided. comprising. a diffuser {hat diffuses light representing the scene and that has a scattering function that is independent of aperture coordinates, a sensor that receives diffused light representing the scene and generates data representing an image; and a hardware processes that uses a point spread function to deblur the image.

Abstract: Methods, systems, and media for swapping faces in images are provided. In some embodiments, a detected face and face data corresponding to an input image is received. A pose bin associated with the detected face is then identified based on the face data. Next, the detected face is aligned to a generic face associated with the pose bin. At least a portion of a candidate face associated with the pose bin is selected. The at least a portion of the candidate face is then copied to a copy of the input image that is aligned with the generic image to form a swapped-face image. The swapped-face image is next aligned to the input image to form an output image, and then the output image is outputted to a display.

Abstract: Methods and systems for coded rolling shutter are provided. In accordance with some embodiments, methods and system are provided that control the readout timing and exposure length for each row of a pixel array in an image sensor, thereby flexibly sampling the three-dimensional space-time value of a scene and capturing sub-images that effectively encode motion and dynamic range information within a single captured image.

Abstract: A multispectral imaging method and system is provided. A multispectral imaging method of the present invention includes determining an on-off combination of a plurality of light sources illuminating a scene; illuminating the scene with the light sources according to the on-off combination selected on the basis of a first control signal generated by a microcontroller; capturing an image of the scene by operating a camera on the basis of a second control signal synchronized with the first control signal; determining a plurality of spectral basis functions and weights of the spectral basis functions; and acquiring a continuous spectral reflectance by summing values obtained by multiplying the spectral basis functions and respective weights. The multispectral imaging method of the present invention is practical and efficient in that a continuous spectral reflectance image can be acquired with a minimized number of measurements required for obtaining spectral reflectance.

Abstract: Embodiments of the present invention provide a lensless optical device for acquiring an image. The device can include a light attenuating layer having a plurality of elements, where transmittance of each of the plurality of elements is controllable, and an image detector disposed at a distance from the light attenuating layer, the image detector configured to acquire an image with light that passes through the light attenuating layer. The device also can include a light attenuating layer controller configured to simultaneously control transmittance of each of the plurality of elements independent of each other. Methods of detecting and tracking an object in a scene are also disclosed.

Abstract: Systems and methods for displaying three-dimensional (3D) images are described. In particular, the systems can include a display block made from a transparent material with optical elements three-dimensionally disposed therein. Each optical element becomes luminous when illuminated by a light ray. The systems can also include a computing device configured to generate two-dimensional (2D) images formatted to create 3D images when projected on the display block, by a video projector coupled to the computing device. The video projector is configured to project the 2D images on the block to create the 3D images by causing a set of the passive optical elements to become luminous. Various other systems and methods are described for displaying 3D images.

Abstract: Systems and methods for providing a substantially de-blurred image of a scene from a motion blurred image of the scene are disclosed. An exemplary system includes a primary detector for sensing the motion blurred image and generating primary image information representing the blurred image, a secondary detector for sensing two or more secondary images of the scene and for generating secondary image information representing the two or more secondary images, and a processor for determining motion information from the secondary image information, estimating a point spread function for the motion blurred image from the motion information, and applying the estimated point spread function to the primary image information to generate information representing the substantially de-blurred image.

Abstract: Methods, systems, and media for automatically classifying face images are provided. In some embodiments, features of the face image to be classified for an attribute are selected, wherein each of the features corresponds to a different region of the face image and specifies one or more of a type of pixel data to be evaluated for the region, a normalization to be applied for the region, and an aggregation to be applied for the region. The face image is classified with respect to the attribute based on the features of the image, and the attribute and a confidence value are assigned to the face image based on the classifying. A query is received from a user, and the attribute is identified as corresponding to the query. The face image is determined as corresponding to the attribute, and the face image is identified to the user as corresponding to the query.

Abstract: A system (100) for the adaptive imaging of a scene includes a digital light processing apparatus (150) adapted for controllably reflecting an image of the scene in at least a first direction to thereby reflect an intensity modulated image of the scene along at least the first direction, an image detector (140) for detecting the intensity modulated image of the scene and generating corresponding image data, and an image data processor (154) for processing the image data into control data and providing the control data to the digital light processing (150) apparatus for control thereof.

Abstract: Systems and methods for panoramic imaging are disclosed herein. An exemplary system of the disclosed subject matter for panoramic imaging includes a fisheye lens, a mirror optically coupled to the fisheye lens, and a detector for capturing light incident on the fisheye lens and reflected from the mirror. In some embodiments, the mirror is a spherical mirror. In some embodiments, the mirror is positioned coaxially with regard to the fisheye lens.

Abstract: Generalized assorted pixel camera systems and methods are provided. In accordance with some embodiments, the generalized assorted pixel camera systems include a color filter array, where the color filter array includes a plurality of primary filters and a plurality of secondary filters. Each filter has a particular spectral response and each filter is formed on a corresponding pixel of a plurality of pixels. Each of the plurality of primary filters and the plurality of secondary filters enhances an attribute of image quality and the information obtained using the plurality of primary filters and the plurality of secondary filters is used to balance spectral resolution, dynamic range, and spatial resolution for generating an image of a plurality of image types.

Abstract: Systems and methods for providing a substantially de-blurred image of a scene from a motion blurred image of the scene are disclosed. An exemplary system includes a primary detector for sensing the motion blurred image and generating primary image information representing the blurred image, a secondary detector for sensing two or more secondary images of the scene and for generating secondary image information representing the two or more secondary images, and a processor for determining motion information from the secondary image information, estimating a point spread function for the motion blurred image from the motion information, and applying the estimated point spread function to the primary image information to generate information representing the substantially de-blurred image.

Abstract: The application provides techniques for obtaining a relatively high dynamic range image of a scene using a relatively low dynamic range image sensor exposed to incident light from the scene for capturing an image. The image sensor has a multiplicity of light-sensing elements in an array and each light sensing element has a particular one of a plurality of sensitivity levels to incident light in accordance with a predetermined sensitivity pattern for the array of light-sensing elements and has a response function. Each light sensing element is responsive to incident light from the scene for producing a captured image brightness value at a corresponding one of a multiplicity of pixel positions of a pixel position array. Each one of the multiplicity of pixel positions corresponds to a particular one of the plurality of sensitivity levels of the light sensing elements.

Abstract: Embodiments of the present invention provide a lensless optical device for acquiring an image. The device can include a light attenuating layer having a plurality of elements, where transmittance of each of the plurality of elements is controllable, and an image detector disposed at a distance from the light attenuating layer, the image detector configured to acquire an image with light that passes through the light attenuating layer. The device also can include a light attenuating layer controller configured to simultaneously control transmittance of each of the plurality of elements independent of each other. Methods of detecting and tracking an object in a scene are also disclosed.

Abstract: Apparatus and methods are provided for obtaining high dynamic range images using a low dynamic range image sensor. The scene is exposed to the image sensor in a spatially varying manner. A variable-transmittance mask, which is interposed between the scene and the image sensor, imposes a spatially varying attenuation on the scene light incident on the image sensor. The mask includes light transmitting cells whose transmittance is controlled by application of suitable control signals. The mask is configured to generate a spatially varying light attenuation pattern across the image sensor. The image frame sensed by the image sensor is normalized with respect to the spatially varying light attenuation pattern. The normalized image data can be interpolated to account for image sensor pixels that are either under or over exposed to enhance the dynamic range of the image sensor.

Abstract: Methods, systems, and media for swapping faces in images are provided. In some embodiments, a detected face and face data corresponding to an input image is received. A pose bin associated with the detected face is then identified based on the face data. Next, the detected face is aligned to a generic face associated with the pose bin. At least a portion of a candidate face associated with the pose bin is selected. The at least a portion of the candidate face is then copied to a copy of the input image that is aligned with the generic image to form a swapped-face image. The swapped-face image is next aligned to the input image to form an output image, and then the output image is outputted to a display.