Atombeam Technologies Inc.

Adeia Guides Inc.

Rovi Guides, Inc.

Advanced New Technologies Co., Ltd.

Zhu Li

Paras Maharjan

Brian R Galvin

Tao Chen

Joshua Cooper

Charles Yeomans

Reza R Derakhshani

Ajita Rattani

Raghunath Sai Puttagunta

Ahmad Saeed Mohammad

Duc Huy Hoang Nguyen

Aliasghar Riahi

A system and method for complex-valued radar image compression integrates AI-based techniques to enhance compression quality. It incorporates a novel AI deblocking network composed of convolutional layers for feature extraction and a channel-wise transformer with attention to capture complex inter-channel dependencies. The convolutional layers extract multi-dimensional features from the complex-valued radar image, while the channel-wise transformer learns global inter-channel relationships. This hybrid approach addresses both local and global features, mitigating compression artifacts and improving image quality. The model's outputs enable effective complex-valued radar image reconstruction, achieving advanced compression while preserving crucial information for accurate analysis.

System and methods for upsampling of decompressed data after lossy compression using a neural network

Systems and methods are provided for predictive mesh coding based on a centroid-normal (C-N) representation. An encoder generates C-N representations of a high-resolution (hi-res) mesh and a downscaling of the mesh (lo-res mesh), each representation having respective centroids and normals. The encoder generates predicted centroids corresponding to the hi-res mesh based on the lo-res centroids using a centroid prediction model. The encoder generates predicted normals corresponding to the hi-res mesh based on the predicted centroids and lo-res normals using a normal vector prediction model. Residuals are computed for the respective predicted geometry data. The encoder transmits encodings of the lo-res mesh and the residuals for decoding at a client device.

Systems and methods for mesh geometry prediction based on a centroid-normal representation

A system and method for Light Detection and Ranging (LIDAR) image compression integrates AI-based techniques to enhance compression quality. It incorporates a novel AI deblocking network composed of convolutional layers for feature extraction and a channel-wise transformer with attention to capture complex inter-channel dependencies. The convolutional layers extract multi-dimensional features from the LIDAR, while the channel-wise transformer learns global inter-channel relationships. This hybrid approach addresses both local and global features, mitigating compression artifacts and improving image quality. The model's outputs enable effective LIDAR image reconstruction, achieving advanced compression while preserving crucial information for accurate analysis.

A system for recovering information lost during genomic data compression employs a quality-driven approach using neural networks. The system evaluates the importance of genomic regions through a quality analysis engine that assigns quality scores, while a rate control engine determines optimal compression rates based on these scores. A specialized neural network recovers lost information from correlated genomic datasets that have undergone lossy compression, utilizing recurrent layers for feature extraction and a channel-wise transformer with attention to capture complex relationships between data channels. The neural network architecture incorporates a deblocking network that combines these components to effectively reconstruct compressed data. A decoder receives and decompresses the data, then processes it through the neural network to recover information lost during compression. This adaptive system ensures critical genomic information is preserved while maximizing compression efficiency.

System and method for adaptive quality driven compression of genomic data using neural networks

A system and method for real time discrete cosine transform image and video processing with convolutional neural network architecture. The system and method transforms degraded inputs into subband images, which are analyzed by a machine learning classification network to identify specific types of blur and compression artifacts. The classification network dynamically adjusts the parameters of separate DC and AC deblurring networks based on its analysis. This adaptive approach optimizes processing for various degradation types, improving the quality of the reconstructed output. The system and method's real-time capability and enhanced adaptability make it suitable for a wide range of imaging and video applications, offering superior performance over traditional methods.

Adaptive real time discrete cosine transform image and video processing with convolutional neural network architecture

A system and method for federated two-stage compression with federated joint learning. The system and method proposed allow for fast and efficient lossless data compression of a large variety of data types. The system and method have a variety of real-world applications, including deep learning solutions for telemetry, tracking, and command subsystems for satellites. Satellites and their control centers are incredibly spaced apart which makes data compression an extremely important process to transmit large sets of information in a low-latency, high-efficiency environment. The proposed system and method utilize probability prediction driven arithmetic long short-term memory system for data compression.

System and method for federated two-stage compression within a persistent cognitive machine

Systems, methods and apparatuses are described herein for training a machine learning model to accept as input synthetic aperture image (SAI) training data for a three-dimensional (3D) display, the 3D display comprising a plurality of layers. The machine learning model may be trained to output respective pixel representations of the SAI training data for each of the plurality of layers of the 3D display. The provided systems, methods and apparatuses may access image data, input the image data to the trained machine learning model, and determine, using the trained machine learning model, respective pixel representations of the input image data for each of the plurality of layers of the 3D display. The provided systems, methods and apparatuses may encode the respective pixel representations of the input image data, and transmit, for display at the 3D display, the encoded respective pixel representations of the input image data.

Learning-based light field compression for tensor display

A system and method are disclosed for adaptive low-light image enhancement using AI-based techniques and wavelet decomposition. The system analyzes characteristics of a raw input image, including brightness, contrast, noise levels, and detail complexity. Based on this analysis, an adaptation control subsystem determines optimal preprocessing parameters. These parameters guide an adaptive wavelet decomposition process, creating multiple frequency domain subimages. Each subimage is processed by a specialized neural network, tailored to its frequency range. The outputs of these networks are combined using an inverse wavelet transform, producing a denoised and enhanced image. This image then undergoes final processing in an image signal processing pipeline. The adaptive nature of the system allows for optimized enhancement across a wide range of low-light conditions, effectively balancing noise reduction, detail preservation, and overall image quality improvement.

System and methods for low-light image enhancement utilizing denoising preprocessing with wavelet decomposition

For compressing synthetic aperture radar (SAR) images, preprocessing operations are performed on an input SAR image. A discrete cosine transform is performed on the image, and multiple subbands are created, where each subband represents a particular range of frequencies. The subbands are organized into multiple groups, where the multiple groups comprise a first low frequency group, a second low frequency group, and a high frequency group. A latent space representation is generated corresponding to each of the multiple groups of subbands. Compression is performed on one or more of the subbands, thereby creating a compressed bitstream, wherein the compressed bitstream is a compressed version of the input SAR image.

Systems and methods for synthetic aperture radar image compression

A system for distributed data compression leverages edge and central computing devices to optimize efficiency and quality. The system includes a lightweight compression subsystem at the edge device that preprocesses and partially compresses input data before transmitting it to a central computing device. The central compression subsystem further processes the data using adjustable compression parameters and a temporal modeling component, reconstructing the data with high fidelity. The system dynamically optimizes preprocessing, encoding, and reconstruction operations based on resource availability, network conditions, and application-specific criteria. This distributed approach improves bandwidth efficiency and adaptability, making it suitable for applications requiring low-latency or resource-constrained environments.

Distributed system and method for adaptive neural network-based data compression

A system and method for federated two-stage compression with federated joint learning. The system and method proposed allow for fast and efficient lossless data compression of a large variety of data types. The system and method have a variety of real-world applications, including deep learning solutions for telemetry, tracking, and command subsystems for satellites. Satellites and their control centers are incredibly spaced apart which makes data compression an extremely important process to transmit large sets of information in a low-latency, high-efficiency environment. The proposed system and method utilize probability prediction driven arithmetic coding which provides faster encoding times and higher compression ratios when paired with a long short-term memory system for data compression.

System and method for federated two-stage compression with federated joint learning

A system and method are disclosed for generating hyperspectral images from RGB (red-green-blue) images. A set of data includes training hyperspectral images and their corresponding RGB images. A spectral band grouping is performed on the training hyperspectral images based on a correlation coefficient of spectral bands. A decomposition network generates a reconstructed hyperspectral image, while a fine-tuning network creates reconstructed RGB images. A quality assurance system analyzes spectral consistency, noise levels, and RGB reconstruction accuracy to generate quality metrics. These metrics guide network weight adjustments to improve reconstruction accuracy.

System and method for hyperspectral image generation with quality assurance

Systems, methods and apparatuses are described herein for accessing image data, generated at least in part using a device comprising a lenslet array, determining a plurality of reference pixel blocks of the image data, and determining a prediction block in a vicinity of the reference pixel blocks. Implementing any of the technique(s) described herein, the system or systems may determine, based on the plurality of reference pixel blocks, a first component representing average pixel values of the prediction block, a second component representing low frequency pixel values of the prediction block, and a third component representing high frequency pixel values of the prediction block. The system(s) may determine predicted pixel values of the prediction block based on the first component, the second component and the third component, and encode the image data based at least in part on the predicted pixel values of the prediction block.

Predictive coding of lenslet images

Systems and methods are described for generating pixel image data, using a lensless camera, based on light that travels through a mask that with pattern masking the lensless camera. The system applies a transformation function to the pixel image data to generate frequency domain image data. The system inputs the frequency domain image data into a machine learning model, wherein the machine learning model does not have access to data that represents the pattern of the mask. The model is trained using a set of images with the feature that are captured by the flat, lensless camera through the mask. The system processes the frequency domain image data using the machine learning model to determine whether the pixel image data depicts the image feature. The system further performs an action based on determining that the pixel image data depicts the image feature.

Growing community of inventors

Overland Park, KS, United States of America

Zhu Li