Patents by Inventor Wenhan Zhang

Wenhan Zhang has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).

  • Publication number: 20210097290
    Abstract: A video retrieval system may include a feature extractor configured to extract first feature descriptors for multiple image frames in the query video. The system may also include a feature extractor to extract second feature descriptors for multiple image frames in a candidate video in a video database. The system may include a comparator to compare the first and second feature descriptors to determine a subset of image frames in the candidate video that are similar to the first video. The system may output die query output by displaying the subset of image frames in a slide show. The system may also output the query by displaying a video formed by at least the subset of image frames. The feature extractor may be implemented in a convolution neural network (CNN) in an artificial intelligence (AI) chip. The system may include key frame extractor to detect key frames in the video.
    Type: Application
    Filed: September 27, 2019
    Publication date: April 1, 2021
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Bin Yang, Qi Dong, Xiaochun Li, Wenhan Zhang, Yequn Zhang, Hua Zhou, Patrick Dong
  • Publication number: 20210019606
    Abstract: An integrated circuit may include multiple cellular neural networks (CNN) processing engines coupled to at least one input/output data bus and a clock-skew circuit in a loop circuit. Each CNN processing engine includes multiple convolution layers, a first memory buffer to store imagery data and a second memory buffer to store filter coefficients. Each of the CNN processing engines is configured to perform convolution operations over an input image simultaneously in a first clock cycle to generate output to be fed to an immediate neighbor CNN processing engine for performing convolution operations in a next clock cycle. The second memory buffer may store a first subset of filter coefficients for a first convolution layer of the CNN processing engine and store a reference location to the first subset of filter coefficients for a second convolution layer, where the filter coefficients for the first and second convolution layers are duplicate.
    Type: Application
    Filed: July 18, 2019
    Publication date: January 21, 2021
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Baohua Sun, Yongxiong Ren, Wenhan Zhang
  • Publication number: 20210019602
    Abstract: An integrated circuit may include multiple cellular neural networks (CNN) processing engines coupled in a loop circuit and configured to perform an AI task. Each CNN processing engine includes multiple convolution layers, a first memory buffer to store imagery data and a second memory buffer to store filter coefficients. The CNN processing engines are configured to perform convolution operations over an input image simultaneously in one or more iterations. In each iteration, various sub-images of the input image are loaded to the first memory buffer circularly. A portion of the filter coefficients corresponding to the sub-image are loaded to the second memory buffer in a cyclic order. Data may be arranged in the second memory buffer to facilitate loading of duplicate filter coefficients among at least two convolution layers without requiring duplicate memory space. Methods of training a CNN model having duplicate weights are also provided.
    Type: Application
    Filed: July 18, 2019
    Publication date: January 21, 2021
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Baohua Sun, Yongxiong Ren, Wenhan Zhang
  • Publication number: 20200392147
    Abstract: Herein are described the preparation of a series of synthetic rocaglates, amidino- and amino-rocaglates, which display inhibition of protein translation and tumor cell proliferation (in vitro and in vivo). The methods described herein allow the preparation of libraries of modified rocaglates. This chemical modification of the rocaglate scaffold changes the C8b-hydroxyl of the natural product series to a more optimal hydrogen bond donor/acceptor.
    Type: Application
    Filed: June 11, 2020
    Publication date: December 17, 2020
    Applicant: TRUSTEES OF BOSTON UNIVERSITY
    Inventors: John A. Porco, Jr., Wenhan Zhang, Jerry Pelletier, Jennifer Chu
  • Publication number: 20200380263
    Abstract: A system for detecting key frames in a video may include a feature extractor configured to extract feature descriptors for each of the multiple image frames in the video. The feature extractor may be an embedded cellular neural network of an artificial intelligence (AI) chip. The system may also include a key frame extractor configured to determine one or more key frames in the multiple image frames based on the corresponding feature descriptors of the image frames. The key frame extractor may determine the key frames based on distance values between a first set of feature descriptors corresponding to a first subset of image frames and a second set of feature descriptors corresponding to a second subset of image frames. The system may output an alert based on determining the key frames and/or display the key frames. The system may also compress the video by removing the non-key frames.
    Type: Application
    Filed: May 29, 2019
    Publication date: December 3, 2020
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Bin Yang, Qi Dong, Xiaochun Li, Wenhan Zhang, Yinbo Shi, Yequn Zhang
  • Publication number: 20200372618
    Abstract: A method of video deblurring by an electronic device is described. The processing circuitry of the electronic device acquires N continuous image frames from a video clip the N being a positive integer, and the N continuous image frames including a blurry image frame to be processed. The processing circuitry of the electronic device performs three-dimensional (3D) convolution processing on the N continuous image frames with a generative adversarial network model, to acquire spatio-temporal information corresponding to the blurry image frame.
    Type: Application
    Filed: August 14, 2020
    Publication date: November 26, 2020
    Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventors: Kaihao ZHANG, Wenhan Luo, Lin Ma, Wei Liu
  • Publication number: 20200320284
    Abstract: Provided is a video processing method, including: obtaining a to-be-processed video and generating a first gait energy diagram, the to-be-processed video including an object with a to-be-recognized identity; obtaining a second gait energy diagram, the second gait energy diagram being generated based on a video including an object with a known identity; inputting the first gait energy diagram and the second gait energy diagram into a deep neural network; extracting respective identity information of the first gait energy diagram and the second gait energy diagram, and determining a fused gait feature vector from gait feature vectors of the first gait energy diagram and the second gait energy diagram; and calculating a similarity based on at least the fused gait feature vector. The identity information of the first gait energy diagram includes gait feature vectors, and the identity information of the second gait energy diagram includes gait feature vectors.
    Type: Application
    Filed: June 17, 2020
    Publication date: October 8, 2020
    Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LTD
    Inventors: Kaihao ZHANG, Wenhan Luo, Lin Ma, Wei Liu
  • Publication number: 20200320385
    Abstract: A system for training an artificial intelligence (AI) model for an AI chip may include a forward network and a backward propagation network. The AI model may be a convolution neural network (CNN). The forward network may infer the output of the AI chip based on the training data. The backward network may use the output of the AI chip and the ground truth data to train the weights of the AI model. In some examples, the system may train the AI model using a gradient descent method. The system may quantize the weights and update the weights during the training. In some examples, the system may perform a uniform quantization over the weights. The system may also determine the distribution of the weights. If the weight distribution is not symmetric, the system may group the weights and quantize the weights based on the grouping.
    Type: Application
    Filed: April 30, 2019
    Publication date: October 8, 2020
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Baohua Sun, Yongxiong Ren, Wenhan Zhang, Patrick Zeng Dong
  • Publication number: 20200302276
    Abstract: An artificial intelligence (AI) semiconductor having an embedded convolution neural network (CNN) may include a first convolution layer and a second convolution layer, in which the weights of the first layer and the weights of the second layer are quantized in different bit-widths, thus at different compression ratios. In a VGG neural network, the weights of a first group of convolution layers may have a different compression ratio than the weights of a second group of convolution layers. The weights of the CNN may be obtained in a training system including convolution quantization and/or activation quantization. Depending on the compression ratio, the weights of a convolution layer may be trained with or without re-training. An AI task, such as image retrieval, may be implemented in the AI semiconductor having the CNN described above.
    Type: Application
    Filed: September 27, 2019
    Publication date: September 24, 2020
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Lin Yang, Bin Yang, Hua Zhou, Xiaochun Li, Wenhan Zhang, Qi Dong, Yequn Zhang, Yongxiong Ren, Patrick Dong
  • Publication number: 20200302180
    Abstract: An image recognition method is provided, including: obtaining a target video including a target object; extracting a target video frame image from the target video; generating a key point video frame sequence comprised of a plurality of key point video frames according to key point information of the object and a plurality of video frames in the target video; extracting dynamic timing feature information of the key point video frame sequence; extracting static structural feature information of the target video frame image; and recognizing an attribute type corresponding to the target object in the target video according to the dynamic timing feature information of the key point video frame sequence and the static structural feature information of the target video frame image.
    Type: Application
    Filed: June 5, 2020
    Publication date: September 24, 2020
    Inventors: Kaihao ZHANG, Wenhan LUO, Lin MA, Wei LIU
  • Patent number: 10733039
    Abstract: This disclosure relates to testing of integrated artificial intelligence (AI) circuit with embedded memory to improve effective chip yield and to mapping addressable memory segments of the embedded memory to multilayer AI networks at the network level, layer level, parameter level, and bit level based on bit error rate (BER) of the addressable memory segments. The disclosed methods and systems allows for deployment of one or more multilayer AI networks in an AI circuit with sufficient model accuracy even when the embedded memory has an overall BER higher than a preferred overall threshold.
    Type: Grant
    Filed: December 21, 2018
    Date of Patent: August 4, 2020
    Assignee: Gyrfalcon Technology Inc.
    Inventors: Chyu-Jiuh Torng, Daniel H. Liu, Wenhan Zhang, Hualiang Yu
  • Publication number: 20200201697
    Abstract: This disclosure relates to testing of integrated artificial intelligence (AI) circuit with embedded memory to improve effective chip yield and to mapping addressable memory segments of the embedded memory to multilayer AI networks at the network level, layer level, parameter level, and bit level based on bit error rate (BER) of the addressable memory segments. The disclosed methods and systems allows for deployment of one or more multilayer AI networks in an AI circuit with sufficient model accuracy even when the embedded memory has an overall BER higher than a preferred overall threshold.
    Type: Application
    Filed: December 21, 2018
    Publication date: June 25, 2020
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Chyu-Jiuh Torng, Daniel H. LIU, Wenhan Zhang, Hualiang Yu
  • Publication number: 20200193280
    Abstract: This disclosure relates to artificial intelligence (AI) circuits with embedded memory for storing trained AI model parameters. The embedded memory cell structure, device profile, and/or fabrication process are designed to generate binary data access asymmetry and error rate asymmetry between writing binary zeros and binary ones that are adapted to and compatible with a binary data asymmetry of the trained model parameters and/or a bit-inversion tolerance asymmetry of the AI model between binary zeros and ones. The disclosed method and system improves predictive accuracy and memory error tolerance without significantly reducing an overall memory error rate and without relying on memory cell redundancy and error correction codes.
    Type: Application
    Filed: December 12, 2018
    Publication date: June 18, 2020
    Applicant: Gyrfalcon Technology Inc.
    Inventors: Chyu-Jiuh Torng, Hualiang Yu, Wenhan Zhang, Daniel H. Liu
  • Patent number: 10654449
    Abstract: A windscreen wiper control system controls a drive mechanism for at least one windscreen wiper arm to effect a reciprocating movement of the at least one windscreen wiper arm within a wiping range between a first position and a second position. The control system includes a detector that detects, upon activation of the control system, an uncertain position of the windscreen wiper arm within the wiping range but different from the first position. Upon detection of such an uncertain position, the control system returns the windscreen wiper arm to the first position at a predetermined reduced speed, the reduced speed being applied at least in a sub-range of the wiping range in the vicinity of the first position.
    Type: Grant
    Filed: June 5, 2015
    Date of Patent: May 19, 2020
    Inventors: Akira Matsuura, Yasushi Azuma, Haruki Nakamura, Tatsuya Ishikawa, Makoto Kukihara, Tsuyoshi Abe, St├ęphane Boursier, Laurent Takejiro Pascal Ochiai Bonneville, Philippe Frin, Tom Teriierooiterai, Wenhan Zhang
  • Publication number: 20200123170
    Abstract: Described herein are compounds, agents, compositions, and methods related to the treatment of a viral infection (e.g., Hepatitis C viral infection). In particular, the compounds, agents, compositions, and methods described herein inhibit viral entry into a target cell.
    Type: Application
    Filed: October 22, 2019
    Publication date: April 23, 2020
    Applicants: Trustees of Boston University, SRI International
    Inventors: John A. Porco, Jr., Wenhan Zhang, Tony Tianyi Wang, Shufeng Liu
  • Publication number: 20200097479
    Abstract: Computers for enabling an entity such as a robot to build other robots that are deployed into service and have a system for tracking performance through a tracking tree of essentially any width or depth and to receive tracking data for the entire width and depth of the tracking tree. In some embodiments, building the multi-line tracking plan includes transitioning an entity's existing tracking plan that is limited in width, depth or width and depth to a new tracking plan that is unlimited in width and depth. In some embodiments, computers can cause an entity to re-enter the tracking tree, within one or more downlines, thereby providing new tracking positions for the same entity within a single tracking tree. Embodiments may include computer systems that auto-balance volume that is generated from the downlines, such that the volume is placed where it would generate the greatest amount of tracking data for a sponsoring entity.
    Type: Application
    Filed: November 27, 2019
    Publication date: March 26, 2020
    Inventors: Fred William Cooper, Riley Paul Timmer, Wenhan Zhang
  • Patent number: 10585988
    Abstract: Systems, methods, and computer-executable instructions for approximating a softmax layer are disclosed. A small world graph that includes a plurality of nodes is constructed for a vocabulary of a natural language model. A context vector is transformed. The small world graph is searched using the transformed context vector to identify a top-K hypothesis. A distance from the context vector for each of the top-K hypothesis is determined. The distance is transformed to an original inner product space. A softmax distribution is computed for the softmax layer over the inner product space of the top-K hypothesis. The softmax layer is useful for determining a next word in a speech recognition or machine translation.
    Type: Grant
    Filed: June 28, 2018
    Date of Patent: March 10, 2020
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Minjia Zhang, Xiaodong Liu, Wenhan Wang, Jianfeng Gao, Yuxiong He
  • Publication number: 20190377792
    Abstract: Systems, methods, and computer-executable instructions for approximating a softmax layer are disclosed. A small world graph that includes a plurality of nodes is constructed for a vocabulary of a natural language model. A context vector is transformed. The small world graph is searched using the transformed context vector to identify a top-K hypothesis. A distance from the context vector for each of the top-K hypothesis is determined. The distance is transformed to an original inner product space. A softmax distribution is computed for the softmax layer over the inner product space of the top-K hypothesis. The softmax layer is useful for determining a next word in a speech recognition or machine translation.
    Type: Application
    Filed: June 28, 2018
    Publication date: December 12, 2019
    Inventors: Minjia Zhang, Xiaodong Liu, Wenhan Wang, Jianfeng Gao, Yuxiong He
  • Publication number: 20190348062
    Abstract: A system for encoding data in an artificial intelligence (AI) integrated circuit solution may include a processor configured to receive image/voice data and generate a sequence of two-dimensional (2D) arrays each array being shifted from a preceding 2D array in the sequence by a time difference. The system may load the sequence of arrays into an AI integrated circuit, feed each of the 2D arrays in the sequence into a respective channel in an embedded cellular neural network architecture in the AI integrated circuit. The system may generate an image/voice recognition result from the embedded cellular neural network architecture and output the image/voice recognition result. The sequence of 2D arrays in the image recognition may include a sequence of output images. The sequence of 2D arrays in the voice recognition may include 2D frequency-time arrays. Sample data may be encoded in a similar manner for training the cellular neural network.
    Type: Application
    Filed: May 8, 2018
    Publication date: November 14, 2019
    Applicant: GYRFALCON TECHNOLOGY INC.
    Inventors: Xiang Gao, Lin Yang, Wenhan Zhang
  • Patent number: 10452955
    Abstract: Methods of encoding image data for loading into an artificial intelligence (AI) integrated circuit are provided. The AI integrated circuit may have an embedded cellular neural network for implementing AI tasks based on the loaded image data. An encoding method may apply image splitting, principal component analysis (PCA) or a combination thereof to an input image to generate a plurality of output images. Each output image has a size smaller than the size of the input image. The method may load the output images into the AI chip, execute programming instructions contained in the AI chip to generate an image recognition result based on the at least one of the plurality of output images, and output the image recognition result. The encoding method also trains a convolution neural network (CNN) and loads the weights of the CNN into the AI integrated circuit for implementing the AI tasks.
    Type: Grant
    Filed: January 15, 2018
    Date of Patent: October 22, 2019
    Assignee: Gyrfalcon Technology Inc.
    Inventors: Xiang Gao, Lin Yang, Wenhan Zhang