Mayachitra, Inc.

University of California

Tejaswi Nanjundaswamy

Bangalore S Manjunath

Shivkumar Chandrasekaran

Michael Gene Goebel

Kenneth Rose

Lakshmanan Nataraj

Tajuddin Manhar Mohammed

Chandrakanth Gudavalli

Kumar Bhargav Viswanatha

Emrah Akyol

Systems, devices, methods and instructions are described for detecting GAN generated images. On embodiment involves receiving an images, generating co-occurrence matrices on color channels of the image, generating analysis of the image by using a convolutional neural network trained to analyze image features of the images based on the generated co-occurrence matrices and determining whether the image is a GAN generated image based on the analysis.

Detecting digital human face image manipulations

Systems and methods herein describe a metadata verification system that is configured to access a digital satellite image, generate a first discrete Fourier transform (DFT) pattern based on an ortho-rectified digital satellite image, generate a second DFT pattern based on rational polynomial coefficient (RPC) data of the digital satellite image, compare the first DFT pattern to the second DFT pattern, generate a score based on the comparison, and generate a determination of whether the digital satellite image has been manipulated based on the score.

Rational polynomial coefficient based metadata verification

Detecting digital image manipulations

A method, device, and apparatus provide the ability to predict a portion of a polyphonic audio signal for compression and networking applications. The solution involves a framework of a cascade of long term prediction filters, which by design is tailored to account for all periodic components present in a polyphonic signal. This framework is complemented with a design method to optimize the system parameters. Specialization may include specific techniques for coding and networking scenarios, where the potential of each enhanced prediction is realized to considerably improve the overall system performance for that application. One specific technique provides enhanced inter-frame prediction for the compression of polyphonic audio signals, particularly at low delay. Another specific technique provides improved frame loss concealment capabilities to combat packet loss in audio communications.

Method and apparatus for polyphonic audio signal prediction in coding and networking systems

A method and apparatus provide the ability to code signals in a layered manner for compression and networking applications. The solution involves a relaxed hierarchical structure of layers, wherein only an optimal subset of information from lower quality levels is transmitted to higher level decoders. This framework is complemented with a design method to optimize system parameters. Specialization may include techniques for employing irregular quantizers and/or estimation theoretic optimal parameter selection and/or content specific optimization (e.g., exploiting harmonic structure in audio, adaptive transform coding and enhanced motion compensated prediction for video) and/or optimization of the structure of the layers, where the potential of exploiting all the common information is realized to improve overall system performance for that application. One specific technique provides improved compression of signals with multiple quality levels with or without prediction. Another specific technique provides improved compression of signals with multiple heterogeneous quality-levels with or without prediction.

Method and apparatus for layered compression of multimedia signals for storage and transmission over heterogeneous networks

Growing community of inventors

Goleta, CA, United States of America

Tejaswi Nanjundaswamy