Snap Inc.

William Marsh Rice University

Carnegie Mellon University

Google Inc.

Northwestern University

Yicheng Wu

Gurunandan Krishnan Gorumkonda

Ashok Veeraraghavan

Aswin C Sankaranarayanan

Huaijin Chen

Vivek Boominathan

Shree Kumar Nayar

Jonathan Tilton Barron

Georgios Evangelidis

Rahul Garg

Oliver Strider Cossairt

Jiawen Chen

Tianfan Xue

Ramzi Zahreddine

Jian Wang

Qiurui He

Sizhuo Ma

Numair Khalil Ullah Khan

Brevin Tilmon

Manoj Kumar Sharma

Jason Holloway

A device that measures a size of a user's face, referred to as face scaling, using a monocular camera. Depth is calculated from sparse feature points. A face mesh is used to improve the estimation accuracy. A processing pipeline detects face features by applying a face landmark detection algorithm to find the important face feature points such as the eyes, nose, and mouth. The processing pipeline estimates feature points depth using depth obtained through image defocus. The processing pipeline further scales the face using an estimated depth of the face features.

Monocular camera defocus face measuring

Images of a scene are received. The images represent viewpoints corresponding to the scene. A pixel map of the scene is computed based on the plurality of images. Multi-plane image (MPI) layers from the pixel map are extracted in real-time. The MPI layers are aggregated. The scene is rendered from a novel viewpoint based on the aggregated MPI layers.

High-definition real-time view synthesis

A method for a passive single-viewpoint 3D imaging system comprises capturing an image from a camera having one or more phase masks. The method further includes using a reconstruction algorithm, for estimation of a 3D or depth image.

Passive and single-viewpoint 3D imaging system

A method includes obtaining an input image that contains a particular representation of lens flare, and processing the input image by a machine learning model to generate a de-flared image that includes the input image with at least part of the particular representation of lens flare removed. The machine learning (ML) model may be trained by generating training images that combine respective baseline images with corresponding lens flare images. For each respective training image, a modified image may be determined by processing the respective training image by the ML model, and a loss value may be determined based on a loss function comparing the modified image to a corresponding baseline image used to generate the respective training image. Parameters of the ML model may be adjusted based on the loss value determined for each respective training image and the loss function.

Learning-based lens flare removal

An energy-efficient adaptive 3D sensing system. The adaptive 3D sensing system includes one or more cameras and one or more projectors. The adaptive 3D sensing system captures images of a real-world scene using the one or more cameras and computes depth estimates and depth estimate confidence values for pixels of the images. The adaptive 3D sensing system computes an attention mask based on the one or more depth estimate confidence values and commands the one or more projectors to send a distributed laser beam into one or more areas of the real-world scene based on the attention mask. The adaptive 3D sensing system captures 3D sensing image data of the one or more areas of the real-world scene and generates 3D sensing data for the real-world scene based on the 3D sensing image data.

Energy-efficient adaptive 3D sensing

A method for imaging objects includes illuminating an object with a light source of an imaging device, and receiving an illumination field reflected by the object. An aperture field that intercepts a pupil of the imaging device is an optical propagation of the illumination field at an aperture plane. The method includes receiving a portion of the aperture field onto a camera sensor, and receiving a sensor field of optical intensity. The method also includes iteratively centering the camera focus along the Fourier plane at different locations to produce a series of sensor fields and stitching together the sensor fields in the Fourier domain to generate an image. The method also includes determining a plurality of phase information for each sensor field in the series of sensor fields, applying the plurality of phase information to the image, receiving a plurality of illumination fields reflected by the object, and denoising the intensity of plurality of illumination fields using Fourier ptychography.

Synthetic apertures for long-range, sub-diffraction limited visible imaging using fourier ptychography

Growing community of inventors

Houston, TX, United States of America

Yicheng Wu