Samsung Electronics Co., Ltd.

Other

John Seokjun Lee

Hamid Rahim Sheikh

Yingmao Li

Gyeongmin Choe

Youngjun Yoo

Michael Oliver Polley

Chenchi Luo

Jong Young Lee

Yibo Xu

Mary S Lee

Weidi Liu

Devendra Kumar Jangid

Manish Goel

Youngmin Kim

John W Glotzbach

Yuming Zhu

Zeeshan Nadir

Hyejung Kim

Pavan Chennagiri

Abhiram Gnanasambandam

Jun Ki Cho

Seung-Chul Jeon

A method for operating an electronic device includes capturing, by an image sensor module, a stream from a pixel array. The method also includes processing, by the image sensor module, the stream to generate a preview stream and a full frame stream. The method further includes compressing, by the image sensor module, the preview stream using a first compression and the full frame stream using a second compression. In addition, the method includes outputting, by the image sensor module, the compressed preview stream and the compressed full frame stream.

Apparatus and method for operating multiple cameras for digital photography

A method includes obtaining an under-display camera (UDC) image captured using a camera located under a display. The method also includes processing, using at least one processing device of an electronic device, the UDC image based on a machine learning model to restore the UDC image. The method further includes displaying or storing the restored image corresponding to the UDC image. The machine learning model is trained using (i) a ground truth image and (ii) a synthetic image generated using the ground truth image and a point spread function that is based on an optical transmission model of the display.

Machine learning model training using synthetic data for under-display camera (UDC) image restoration

A method includes identifying, using at least one processing device of an electronic device, a spatially-variant point spread function associated with an under-display camera. The spatially-variant point spread function is based on an optical transmission model and a layout of a display associated with the under-display camera. The method also includes generating, using the at least one processing device, a ground truth image. The method further includes performing, using the at least one processing device, a convolution of the ground truth image based on the spatially-variant point spread function in order to generate a synthetic sensor image. The synthetic sensor image represents a simulated image captured by the under-display camera. In addition, the method includes providing, using the at least one processing device, the synthetic sensor image and the ground truth image as an image pair to train a machine learning model to perform under-display camera point spread function inversion.

A method includes extracting multiple shallow features from a low-resolution image using a shallow feature extractor that includes a quaternion convolutional network. The method also includes extracting multiple deep features from the multiple shallow features using a deep feature extractor that includes multiple quaternion residual distillation blocks (QRDBs), where each QRDB includes a quaternion self-attention module. The method further includes reconstructing the multiple deep features into a high-resolution image. Each QRDB may further include a quaternion gated deconvolutional feed forward network (QGDFN) configured to suppress one or more of the multiple deep features.

System and method for single image super- resolution for smart device camera

A method includes obtaining an input image that contains blur. The method also includes providing the input image to a trained machine learning model, where the trained machine learning model includes (i) a shallow feature extractor configured to extract one or more feature maps from the input image and (ii) a deep feature extractor configured to extract deep features from the one or more feature maps. The method further includes using the trained machine learning model to generate a sharpened output image. The trained machine learning model is trained using ground truth training images and input training images, where the input training images include versions of the ground truth training images with blur created using demosaic and noise filtering operations.

Machine learning-based approaches for synthetic training data generation and image sharpening

An apparatus includes at least one processing device configured to obtain input frames from a video. The at least one processing device is also configured to generate a forward flow from a first input frame to a second input frame and a backward flow from the second input frame to the first input frame. The at least one processing device is further configured to generate an occlusion map at an interpolated frame coordinate using the forward flow and the backward flow. The at least one processing device is also configured to generate a consistency map at the interpolated frame coordinate using the forward flow and the backward flow. In addition, the at least one processing device is configured to perform blending using the occlusion map and the consistency map to generate an interpolated frame at the interpolated frame coordinate.

Pixel blending for synthesizing video frames with occlusion and watermark handling

A method includes obtaining multiple video frames. The method also includes determining whether a bi-directional optical flow between the multiple video frames satisfies an image quality criterion for bi-directional consistency. The method further includes identifying a non-linear curve based on pixel coordinate values from at least two of the video frames. The at least two video frames include first and second video frames. The method also includes generating interpolated video frames between the first and second video frames by applying non-linear interpolation based on the non-linear curve. In addition, the method includes outputting the interpolated video frames for presentation.

Accurate optical flow interpolation optimizing bi-directional consistency and temporal smoothness

A method includes obtaining a first optical flow vector representing motion between consecutive video frames during a previous time step. The method also includes generating a first predicted optical flow vector from the first optical flow vector using a trained prediction model, where the first predicted optical flow vector represents predicted motion during a current time step. The method further includes refining the first predicted optical flow vector using a trained update model to generate a second optical flow vector representing motion during the current time step. The trained update model uses the first predicted optical flow vector, a video frame of the previous time step, and a video frame of the current time step to generate the second optical flow vector.

Multi-frame optical flow network with lossless pyramid micro-architecture

A method includes obtaining, using a stationary sensor of an electronic device, multiple image frames including first and second image frames. The method also includes generating, using multiple previously generated motion vectors, a first motion-distorted image frame using the first image frame and a second motion-distorted image frame using the second image frame. The method further includes adding noise to the motion-distorted image frames to generate first and second noisy motion-distorted image frames. The method also includes performing (i) a first multi-frame processing (MFP) operation to generate a ground truth image using the motion-distorted image frames and (ii) a second MFP operation to generate an input image using the noisy motion-distorted image frames. In addition, the method includes storing the ground truth and input images as an image pair for training an artificial intelligence/machine learning (AI/ML)-based image processing operation for removing image distortions caused by handheld image capture.

Hand motion pattern modeling and motion blur synthesizing techniques

Infinity smart serving robot

An apparatus includes at least one memory configured to store an AI network and at least one processor. The at least one processor is configured to generate a dead leaves model. The at least one processor is also configured to capture a ground truth frame from the dead leaves. The at least one processor is further configured to apply a mathematical noise model to the ground truth frame to produce a noisy frame. In addition, the at least one processor is configured to train the AI network using the ground truth frame and the noisy frame.

Generating a synthetic ground-truth image using a dead leaves model

A method includes obtaining, using at least one processor of an electronic device, multiple calibration parameters associated with multiple sensors of a selected mobile device. The method also includes obtaining, using the at least one processor, an identification of multiple imaging tasks. The method further includes obtaining, using the at least one processor, multiple synthetically-generated scene images. In addition, the method includes generating, using the at least one processor, multiple training images and corresponding meta information based on the calibration parameters, the identification of the imaging tasks, and the scene images. The training images and corresponding meta information are generated concurrently, different ones of the training images correspond to different ones of the sensors, and different pieces of the meta information correspond to different ones of the imaging tasks.

Multi-sensor, multi-view, multi-frame, multi-task synthetic image fusion engine for mobile imaging system

A method, an electronic device, and computer readable medium are provided. The method includes receiving an input into a neural network that includes a kernel. The method also includes generating, during a convolution operation of the neural network, multiple panel matrices based on different portions of the input. The method additionally includes successively combining each of the multiple panel matrices with the kernel to generate an output. Generating the multiple panel matrices can include mapping elements within a moving window of the input onto columns of an indexing matrix, where a size of the window corresponds to the size of the kernel.

Convolution streaming engine for deep neural networks

A method includes, in a first mode, positioning first and second tiltable image sensor modules of an image sensor array of an electronic device so that a first optical axis of the first tiltable image sensor module and a second optical axis of the second tiltable image sensor module are substantially perpendicular to a surface of the electronic device, and the first and second tiltable image sensor modules are within a thickness profile of the electronic device. The method also includes, in a second mode, tilting the first and second tiltable image sensor modules so that the first optical axis of the first tiltable image sensor module and the second optical axis of the second tiltable image sensor module are not perpendicular to the surface of the electronic device, and at least part of the first and second tiltable image sensor modules are no longer within the thickness profile of the electronic device.

A method includes receiving a selection of a selected zoom area on an input image frame displayed on a user interface; determining one or more candidate zoom previews proximate to the selected zoom area using a saliency detecting algorithm; and displaying the one or more candidate zoom previews on the user interface adjacent to the selected zoom area.

Growing community of inventors

Allen, TX, United States of America

John Seokjun Lee