Qualcomm Incorporated

Other

Hau Hwang

Hsiang-Tsun Li

Kalin Mitkov Atanassov

Joseph Cheung

Jisoo Lee

Ruben Manuel Velarde

Vishal I Gupta

Tushar Sinha Pankaj

Sergiu Radu Goma

Shusil Dangi

Milivoje M Aleksic

Szepo Robert Hung

Xiaoyun Jiang

Shizhong Liu

Chinchuan Andrew Chiu

Jingqiang Li

Babak Forutanpour

Khaled Helmi El-Maleh

Ananthapadmanabhan Arasanipalai Kandhadai

Jiafu Luo

Hasib Siddiqui

Venkata Ravi Kiran Dayana

King-Chung Lai

Stephen Horton

Kalin Atanssov

Hasib Ahmed Siddiqui

Hsiang-Tsu Li

Systems and techniques are described for image processing. An imaging system can include an image sensor that captures image data. An image signal processor (ISP) of the imaging system can demosaic the image data. The imaging system can input the image data into one or more trained machine learning models, in some cases along with metadata associated with the image data. The one or more trained machine learning models can output settings for a set of parameters of the ISP based on the image data and/or the metadata. The imaging system can generate an output image by processing the image data using the ISP, with the parameters of the ISP set according to the settings. Each pixel of the pixels of the image data can be processed using a respective setting for adjusting a corresponding parameter. The parameters of the ISP can include gain, offset, gamma, and Gaussian filtering.

Composite image signal processor

Techniques are provided for using one or more machine learning systems to process input data including image data. The input data including the image data can be obtained, and at least one machine learning system can be applied to at least a portion of the image data to determine at least one color component value for one or more pixels of at least the portion of the image data. Based on application of the at least one machine learning system to at least the portion of the image data, output image data for a frame of output image data can be generated. The output image data includes at least one color component value for one or more pixels of the frame of output image data. Application of the at least one machine learning system causes the output image data to have a reduced dimensionality relative to the input data.

Image signal processor for processing images

An imaging system can obtain image data, for instance from an image sensor. The imaging system can supply the image data as input data to a machine learning system, which can generate one or more maps based on the image data. Each map can identify strengths at which a certain image processing function is to be applied to each pixel of the image data. Different maps can be generated for different image processing functions, such as noise reduction, sharpening, or color saturation. The imaging system can generate a modified image based on the image data and the one or more maps, for instance by applying each of one or more image processing functions in accordance with each of the one or more maps. The imaging system can supply the image data and the one or more maps to a second machine learning system to generate the modified image.

Machine learning based image adjustment

Techniques and systems are provided for providing a rendering engine model for raw media data. In some examples, a system obtains media data captured by a data capturing device and embeds, in a media item containing the media data, a rendering engine model including a description of a neural network configured to process the media data and generate a particular media data output, the description defining a neural network architecture for the neural network. The system then outputs the media item with the rendering engine model embedded in the media item, the rendering engine model indicating how to execute the neural network to process the media data in the media item and generate the particular media data output based on the description of the neural network.

Embedded rendering engine for media data

Techniques and systems are provided for processing image data using one or more neural networks. For example, a patch of raw image data can be obtained. The patch can include a subset of pixels of a frame of raw image data, and the frame can be captured using one or more image sensors. The patch of raw image data includes a single color component for each pixel of the subset of pixels. At least one neural network can be applied to the patch of raw image data to determine a plurality of color component values for one or more pixels of the subset of pixels. A patch of output image data can then be generated based on application of the at least one neural network to the patch of raw image data. The patch of output image data includes a subset of pixels of a frame of output image data, and also includes the plurality of color component values for one or more pixels of the subset of pixels of the frame of output image data. Application of the at least one neural network causes the patch of output image data to include fewer pixels than the patch of raw image data. Multiple patches from the frame can be processed by the at least one neural network in order to generate a final output image. In some cases, the patches from the frame can be overlapping so that the final output image contains a complete picture.

Apparatus and methods for conditional display of a stereoscopic image pair on a display device are disclosed. In some aspects, a vertical disparity between two images is corrected. If the corrected vertical disparity is below a threshold, a three dimensional image may be generated based on the correction. In some cases, the corrected vertical disparity may still be significant, for example, above the threshold. In these instances, the disclosed apparatus and methods may display a two dimensional image.

Methods and apparatus for conditional display of a stereoscopic image pair

Apparatus and methods for conditional display of a stereoscopic image pair on a display device are disclosed. Particularly, some implementations include receiving a first image and a second image, determining a vertical disparity between the first image and the second images, and displaying a stereoscopic image pair if the vertical disparity is below a threshold. Some implementations provide for correcting the vertical disparity by generating at least one corrected image, and generating the stereoscopic image pair based on the corrected image. Some implementations may evaluate the quality of the stereoscopic image pair, and display either a two dimensional image or the stereoscopic image pair based on the evaluation.

Methods and apparatus are presented herein to perform selective and/or scalable complexity control of the video codec, so that the amount of processing resources consumed by a video codec may be increased or reduced. Based on the configurable thresholds set within complexity control algorithms, the nonpredictive and the predictive coding sections of the video codec may be selectively implemented. The configurable thresholds are used to determine whether a desired amount of spatial information, such as texture information or motion information, is present within a video frame.

Selective and/or scalable complexity control for video codecs

A method of combining data from multiple sensors is disclosed. The method includes providing a common control signal to multiple image sensors. Each of the multiple image sensors is responsive to the common control signal to generate image data. The method also includes receiving synchronized data output from each of the multiple image sensors.

Combining data from multiple image sensors

This disclosure describes techniques for producing high dynamic range images by applying a variable weighting factor to a sample prior to combining the sample with another sample. In one example, a method includes sampling a first pixel cell signal at a first time to produce a first sample, sampling a second pixel cell signal at a second time to produce a second sample, applying a variable weighting factor to the second sample, wherein the variable weighting factor is defined based on a function, and combining the first sample and the weighted second sample.

High dynamic range image sensor

This disclosure describes techniques for rotating an encoded image, such as an image encoded according to a JPEG standard. In one example, a method for rotating an encoded image comprising reordering minimum coded units (MCUs) of the encoded image according to a specified rotation of the encoded image, rotating image data within the MCUs according to the specified rotation, and generating a rotated version of the encoded image comprising the reordered MCUs and the rotated image data within the MCUs.

Rotational transcoding for JPEG or other coding standards

A method of combining data from multiple sensors is disclosed. The method includes receiving lines of image data at an image processor having an input for a single camera. Each line of the image data includes first line data from a first image captured by a first camera and second line data from a second image captured by a second camera. The method also includes generating an output frame having a first section corresponding to line data of the first image and having a second section corresponding to line data of the second image. The first section and the second section are configured to be used to generate a three-dimensional (3D) image format or a 3D video format.

Growing community of inventors

San Diego, CA, United States of America

Hau Hwang