知搜-全球专利检索

  1. Login
  2. Sign Up

Search Results

10 records (took 0.018 seconds)

    Controllable lossy compression system using joint learning
    1.
    发明授权
    Controllable lossy compression system using joint learning 有权

    公开(公告)号：US12199643B1

    公开(公告)日：2025-01-14

    申请号：US18791425

    申请日：2024-08-01

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Paras Maharjan ,  Brian Galvin

    IPC: H03M7/00 ,  H03M7/30

    Abstract: A system and method for controllable lossy data compression employing a joint learning framework to efficiently compress and reconstruct input data while balancing compression ratio and reconstruction quality. The system comprises an encoding system, a temporal modeling system, and a decoding system, which are jointly optimized to minimize a combined loss function. The encoding system, such as a Vector Quantized Variational Autoencoder (VQ-VAE) compresses the input data into a compact representation, while introducing a controllable degree of lossy compression based on adjustable compression parameters. The temporal modeling system, such as a Multilayer Perceptron Long Short-Term Memory captures temporal dependencies in the compressed representation. The decoding system, such as a VQ-VAE decoder, reconstructs the input data from the compressed representation. By providing control over the trade-off between compression ratio and reconstruction quality, the system offers flexibility for diverse applications.

    Upsampling of compressed financial time-series data using a jointly trained Vector Quantized Variational Autoencoder neural network
    2.
    发明授权
    Upsampling of compressed financial time-series data using a jointly trained Vector Quantized Variational Autoencoder neural network 有权

    公开(公告)号：US12229679B1

    公开(公告)日：2025-02-18

    申请号：US18822203

    申请日：2024-09-01

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Brian Galvin ,  Paras Maharjan

    IPC: G06N3/08 ,  G06N3/0455 ,  G06N3/0495

    Abstract: A system and methods for upsampling compressed data using a jointly trained Vector Quantized Variational Autoencoder (VQ-VAE) and neural upsampler. The system compresses input data into a discrete latent space using a VQ-VAE encoder, reconstructs the data using a VQ-VAE decoder, and enhances the reconstructed data using a neural upsampler. The VQ-VAE and neural upsampler are jointly trained using a combined loss function, enabling end-to-end optimization. The system allows for efficient compression and high-quality reconstruction of various data types, including financial time-series, images, audio, video, sensor data, and text. The learned discrete latent space can be explored and manipulated using techniques such as interpolation, extrapolation, and vector arithmetic to generate new or modified data samples. The system finds applications in data storage, transmission, analysis, and generation across multiple domains.

    Systems and methods for neural network based data compression
    4.
    发明授权
    Systems and methods for neural network based data compression 有权

    公开(公告)号：US12224777B1

    公开(公告)日：2025-02-11

    申请号：US18907442

    申请日：2024-10-04

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Paras Maharjan ,  Brian Galvin

    IPC: H03M7/30

    Abstract: For compressing data, preprocessing operations are performed on raw input data. A discrete cosine transform is performed on the preprocessed data, 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. A first bitstream is created based on the latent space representation, and an alternate representation of the latent space is used for creating a second bitstream, enabling multiple-pass techniques for data compression. The multiple bitstreams may be multiplexed to form a combined bitstream for storage and/or transmission purposes.

    Systems and methods for hyperspectral image generation
    7.
    发明授权
    Systems and methods for hyperspectral image generation 有权

    公开(公告)号：US12190573B1

    公开(公告)日：2025-01-07

    申请号：US18627451

    申请日：2024-04-05

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Paras Maharjan

    IPC: G06V10/58 ,  G06V10/75 ,  G06V10/82

    Abstract: 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 is used to generate a reconstructed hyperspectral image. A fine-tuning network is used to create a reconstructed RGB images. The difference between an input RGB image and a corresponding reconstructed RGB image is used to adjust one or more weights of one or more of the networks.

    System and method for learning-based lossless data compression
    8.
    发明授权
    System and method for learning-based lossless data compression 有权

    公开(公告)号：US12119848B1

    公开(公告)日：2024-10-15

    申请号：US18623018

    申请日：2024-03-31

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Paras Maharjan

    IPC: H03M7/00 ,  H03M7/30

    CPC classification number: H03M7/6011

    Abstract: A system and method learning-based lossless data compression. 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 methods for upsampling of decompressed time-series data using a neural network
    10.
    发明授权
    System and methods for upsampling of decompressed time-series data using a neural network 有权

    公开(公告)号：US12068761B1

    公开(公告)日：2024-08-20

    申请号：US18410980

    申请日：2024-01-11

    Applicant: AtomBeam Technologies Inc.

    Inventor： Zhu Li ,  Brian R. Galvin ,  Paras Maharjan

    IPC: H03M7/30 ,  H04N19/80

    CPC classification number: H03M7/3062 ,  H04N19/80

    Abstract: A system and methods for upsampling of decompressed time-series data after lossy compression using a neural network that integrates AI-based techniques to enhance compression quality. It incorporates a novel AI deblocking network composed of recurrent layers for feature extraction and a channel-wise transformer with attention to capture complex inter-channel dependencies. The recurrent layers extract multi-dimensional features from the two or more correlated datasets, while the channel-wise transformer learns global inter-channel relationships. This hybrid approach addresses both local and global features, mitigating compression artifacts and improving decompressed data quality. The model's outputs enable effective data reconstruction, achieving advanced compression while preserving crucial information for accurate analysis.

Patent Agency Ranking