Patents by Inventor Jianyu Wang
Jianyu Wang 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: 20200410228Abstract: Described herein are embodiments for fast training of adversarially robust models against adversarial attacks. Embodiments for model training by perturbing both the image and the label, which may be referred to as Bilateral Adversarial Training (BAT), are disclosed. To generate the adversarial label, one or more closed-form heuristic solutions are derived. One-step targeted attack is used with the target label being the most confusing class. It is shown in various experiments that random start and the most confusing target attack effectively prevent the label leaking and gradient masking problem. Coupled with the adversarial label part, embodiments of presented models significantly improve the state-of-the-art results. Experiments on one or more computationally challenging dataset also demonstrate the effectiveness of the presented BAT method embodiments.Type: ApplicationFiled: June 28, 2019Publication date: December 31, 2020Applicant: Baidu USA LLCInventors: Jianyu WANG, Haichao ZHANG
-
Publication number: 20200397713Abstract: A chloroquine gel and a preparation method and application thereof. A chloroquine nanosphere includes a water-soluble nanosphere carrier, and chloroquine or a chloroquine derivative, wherein a mass ratio of the chloroquine or the chloroquine derivative to the water-soluble nanosphere carrier is no more than 1:3.Type: ApplicationFiled: November 4, 2019Publication date: December 24, 2020Applicants: GUANGZHOU HYBRIBIO MEDICINE TECHNOLOGY LTD., GUANGZHOU HYBRIBIO BIOTECH LTD.Inventors: Longxu XIE, Xiangling LI, Manli YUAN, Ting WANG, Jianyu WANG
-
Publication number: 20200380300Abstract: Described herein are embodiments for an approach to improve the robustness of an object detector against adversarial attacks. Existing attacks for object detectors and the impacts of individual task component on model robustness are systematically analyzed from a multi-task view of object detection. In one or more embodiments, a multi-task learning perspective of object detection is introduced and an asymmetric role of task losses is identified. One or more embodiments of an adversarial training method for robust object detection are presented to leverage the multiple sources of attacks for improving the robustness of object detection models.Type: ApplicationFiled: May 30, 2019Publication date: December 3, 2020Applicant: Baidu USA LLCInventors: Haichao ZHANG, Jianyu WANG
-
Publication number: 20200265271Abstract: Described herein are embodiments for joint adversarial training methods that incorporate both spatial transformation-based and pixel-value based attacks for improving image model robustness. Embodiments of a spatial transformation-based attack with an explicit notion of budgets are disclosed and embodiments of a practical methodology for efficient spatial attack generation are also disclosed. Furthermore, both pixel and spatial attacks are integrated into embodiments of a generation model and the complementary strengths of each other are leveraged for improving the overall model robustness. Extensive experimental results on several benchmark datasets compared with state-of-the-art methods verified the effectiveness of the presented method.Type: ApplicationFiled: April 30, 2019Publication date: August 20, 2020Applicant: Baidu USA LLCInventors: Haichao ZHANG, Jianyu WANG
-
Publication number: 20190319861Abstract: This disclosure relates to a resource management system of a computing network. The design of the resource management system is an edge-Internet of Things (IoT) framework based on blockchain and smart contracts. The resource management system integrates a permissioned blockchain to link the edge cloud resources with each account for an IoT device, resource usage, and behavior of the IoT device. The resource management system uses credit-based resource management to control the resource of the IoT device that can be obtained from the edge server. Smart contracts are used to regulate behavior of the IoT device and enforce policies.Type: ApplicationFiled: April 12, 2019Publication date: October 17, 2019Inventors: Jianli Pan, Jianyu Wang
-
Patent number: 10074164Abstract: Provided are a method and an apparatus for adjusting image brightness. The method includes: acquiring a single-channel brightness image based on grayscales of each channel of an image to be processed; performing Gaussian filtering on the single-channel brightness image to acquire a Gaussian filtered image; adjusting grayscales of the Gaussian filtered image based on the grayscales of the Gaussian filtered image and a preset proportion; acquiring a grayscale change rate of each pixel with respect to the Gaussian filtered image after adjustment and the Gaussian filtered image before adjustment; and processing the image to be processed based on the grayscale change rate of each pixel to acquire a processed image.Type: GrantFiled: October 17, 2017Date of Patent: September 11, 2018Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventors: Shimin Hu, Jianyu Wang, Baoli Li, Haifeng Deng, Xianying Li
-
Publication number: 20180053290Abstract: Provided are a method and an apparatus for adjusting image brightness. The method includes; acquiring a single-channel brightness image based on grayscales of each channel of an image to be processed; performing Gaussian filtering on the single-channel brightness image to acquire a Gaussian filtered image; adjusting grayscales of the Gaussian filtered image based on the grayscales of the Gaussian filtered image and a preset proportion; acquiring a grayscale change rate of each pixel with respect to the Gaussian filtered image after adjustment and the Gaussian filtered image before adjustment; and processing the image to be processed based on the grayscale change rate of each pixel to acquire a processed image.Type: ApplicationFiled: October 17, 2017Publication date: February 22, 2018Inventors: Shimin HU, Jianyu WANG, Baoli LI, Haifeng DENG, Xianying LI
-
Patent number: 9824430Abstract: Provided are a method and an apparatus for adjusting image brightness. The method includes: acquiring an image to be processed, and acquiring a single-channel brightness image based on grayscales of each channel of the image to be processed; performing Gaussian filtering on the single-channel brightness image to acquire a Gaussian filtered image; adjusting grayscales of the Gaussian filtered image based on the grayscales of the Gaussian filtered image and a preset proportion; comparing the Gaussian filtered image after adjustment with the Gaussian filtered image before adjustment to acquire a grayscale change rate of each pixel through adjustment; processing the image to be processed based on the grayscale change rate of each pixel to acquire a processed image; and outputting the processed image.Type: GrantFiled: July 22, 2015Date of Patent: November 21, 2017Assignee: Tencent Technology (Shenzhen) Company LimitedInventors: Shimin Hu, Jianyu Wang, Baoli Li, Haifeng Deng, Xianying Li
-
Patent number: 9736336Abstract: The disclosure relates to an image tone adjustment method, an apparatus thereof, and a computer-readable storage medium. The method includes: fitting overlapping areas to obtain a first correction parameter and a second correction parameter, performing interpolation on each pixel of the target image to obtain a corresponding correction parameter, correcting the target image according to the corresponding correction parameters; fitting an overlapping area between the corrected target image and the first image, and an overlapping area between the corrected target image and the second image to obtain a first RGB three-channel linear transformation parameter and a second RGB three-channel linear transformation parameter, performing linear interpolation on each pixel of the corrected target image to obtain a corresponding RGB three-channel linear transformation parameter, and transforming the corrected target image.Type: GrantFiled: July 17, 2015Date of Patent: August 15, 2017Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventors: Shimin Hu, Jianyu Wang, Baoli Li, Haifeng Deng, Xianying Li
-
Publication number: 20170149119Abstract: A metal housing of an electronic communication device includes: a metal frame, and a plurality of micro holes. At least some of the plurality of micro holes penetrate the metal frame, with the plurality of micro holes being arranged into at least two rows to form a strip.Type: ApplicationFiled: February 1, 2017Publication date: May 25, 2017Inventors: Jianyu WANG, Zhuxin LI, Guohua WANG
-
Patent number: 9501104Abstract: A screw device, for mounting within a housing, configured to fixedly connect a first portion and a second portion of the housing, includes: a driving part fixedly connected within the first portion of the housing; a screw connected with the driving part; a nut fixedly connected within the second portion of the housing and matching the screw; and a chip configured to control the driving part to drive the screw to move relative to the nut, so as to achieve engagement or separation between the screw and the nut.Type: GrantFiled: September 8, 2014Date of Patent: November 22, 2016Assignee: Xiaomi Inc.Inventors: Yuan Gao, Fenghui Wu, Jianyu Wang
-
Publication number: 20150326753Abstract: The disclosure relates to an image tone adjustment method, an apparatus thereof, and a computer-readable storage medium. The method includes: fitting overlapping areas to obtain a first correction parameter and a second correction parameter, performing interpolation on each pixel of the target image to obtain a corresponding correction parameter, correcting the target image according to the corresponding correction parameters; fitting an overlapping area between the corrected target image and the first image, and an overlapping area between the corrected target image and the second image to obtain a first RGB three-channel linear transformation parameter and a second RGB three-channel linear transformation parameter, performing linear interpolation on each pixel of the corrected target image to obtain a corresponding RGB three-channel linear transformation parameter, and transforming the corrected target image.Type: ApplicationFiled: July 17, 2015Publication date: November 12, 2015Inventors: Shimin HU, Jianyu WANG, Baoli LI, Haifeng DENG, Xianying LI
-
Publication number: 20150324961Abstract: Provided are a method and an apparatus for adjusting image brightness. The method includes: acquiring an image to be processed, and acquiring a single-channel brightness image based on grayscales of each channel of the image to be processed; performing Gaussian filtering on the single-channel brightness image to acquire a Gaussian filtered image; adjusting grayscales of the Gaussian filtered image based on the grayscales of the Gaussian filtered image and a preset proportion; comparing the Gaussian filtered image after adjustment with the Gaussian filtered image before adjustment to acquire a grayscale change rate of each pixel through adjustment; processing the image to be processed based on the grayscale change rate of each pixel to acquire a processed image; and outputting the processed image.Type: ApplicationFiled: July 22, 2015Publication date: November 12, 2015Inventors: Shimin HU, Jianyu WANG, Baoli LI, Haifeng DENG, Xianying LI
-
Patent number: 9165338Abstract: A method for panoramic image completion is disclosed. The method includes: acquiring a panoramic image; obtaining a projected image by mapping pixels of the panoramic image onto a polar coordinate system, wherein a long side component of the pixel coordinate of the pixels is corresponding to the polar angle of the polar coordinate system and a short side component of the pixel coordinate of the pixels is corresponding to a radial coordinate of the polar coordinate system; acquiring an incomplete region of the projected image, and obtaining a completed image by completing the incomplete region; and obtaining a completed panoramic image by inverse mapping the pixels of the completed image according to the polar coordinate system. Furthermore, a device for panoramic image completion is also disclosed. The above method and device for panoramic image completion take account of the perspective curvature of the panoramic image, to improve the degree of restoration.Type: GrantFiled: August 1, 2014Date of Patent: October 20, 2015Assignee: Tencent Technology (Shenzhen) Company LimitedInventors: Kun Xu, Jianyu Wang, Baoli Li, Haifeng Deng, Zhe Zhu
-
Publication number: 20150285649Abstract: A method and an apparatus for determining a traffic route in an electronic map are provided. The method includes: determining by a server that includes a processor, a route start point and a route end point entered through a terminal device that includes a processor; calculating a sectional riding route between the route start point and the route end point in an electronic map road network, setting identification information in the sectional riding route; calculating a sectional walking route corresponding to the sectional riding route in the electronic map road network, integrating the sectional walking route and the sectional riding route into a traffic route according to the identification information, and displaying the traffic route on the terminal device.Type: ApplicationFiled: June 17, 2015Publication date: October 8, 2015Inventors: Xianxin ZHANG, Kang ZHOU, Haibo LU, Jianyu WANG
-
Publication number: 20150170332Abstract: A method for panoramic image completion is disclosed. The method includes: acquiring a panoramic image; obtaining a projected image by mapping pixels of the panoramic image onto a polar coordinate system, wherein a long side component of the pixel coordinate of the pixels is corresponding to the polar angle of the polar coordinate system and a short side component of the pixel coordinate of the pixels is corresponding to a radial coordinate of the polar coordinate system; acquiring an incomplete region of the projected image, and obtaining a completed image by completing the incomplete region; and obtaining a completed panoramic image by inverse mapping the pixels of the completed image according to the polar coordinate system. Furthermore, a device for panoramic image completion is also disclosed. The above method and device for panoramic image completion take account of the perspective curvature of the panoramic image, to improve the degree of restoration.Type: ApplicationFiled: August 1, 2014Publication date: June 18, 2015Applicant: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventors: Kun Xu, Jianyu Wang, Baoli Li, Haifeng Deng, Zhe Zhu
-
Publication number: 20150085440Abstract: A screw device, for mounting within a housing, configured to fixedly connect a first portion and a second portion of the housing, includes: a driving part fixedly connected within the first portion of the housing; a screw connected with the driving part; a nut fixedly connected within the second portion of the housing and matching the screw; and a chip configured to control the driving part to drive the screw to move relative to the nut, so as to achieve engagement or separation between the screw and the nut.Type: ApplicationFiled: September 8, 2014Publication date: March 26, 2015Inventors: Yuan Gao, Fenghui Wu, Jianyu Wang
-
Publication number: 20140240311Abstract: Transitions between street view images are described. The described techniques include: obtaining an original street view image and a target street view image; constructing a three-dimensional model corresponding to the original street view image by three-dimensional modeling; obtaining matching pairs of feature points that are extracted from the original street view image and the target street view image, and simulating a virtual camera in the three-dimensional model according to matching pairs of feature points to capture street view image sequence; and switching from the original street view image to the target street view image according to the street view image sequence. Transition stability is thereby improved.Type: ApplicationFiled: May 1, 2014Publication date: August 28, 2014Applicant: Tencent Technology (Shenzhen) Company LimitedInventors: Kun XU, Jianyu Wang, Baoli Li, Chengjun Li, Haozhi Huang
-
Patent number: 8611644Abstract: Embodiments of the present invention provide a method and apparatus for training an image classifier. The method includes: A. dividing a set of training images for classifier training into a positive-example sample set and at least two negative-example sample sets; B. determining, for each negative-example sample set, a feature set for differentiating the positive-example sample set from the negative-example sample set; and C. performing training using each feature set determined to obtain a classifier. This invention also provides a method and apparatus for image recognition utilizing the image classifier.Type: GrantFiled: August 16, 2010Date of Patent: December 17, 2013Assignee: Tencent Technology (Shenzhen) Company LimitedInventors: Libo Fu, Jianyu Wang, Bo Chen
-
Patent number: 8441485Abstract: The examples of the present invention provide a method and apparatus for generating a video animation, and the method and apparatus relate to the animation field. The method includes: receiving a command sent by a user, determining an action corresponding to the command according to the command, and determining the total number of frames corresponding to the action and a motion coefficient of each frame; calculating an offset of each control point in each frame according to the motion coefficient of each frame, and generating a video animation according to the offset of each control point in each frame and the total number of frames. An apparatus for generating a video animation is also provided.Type: GrantFiled: June 3, 2011Date of Patent: May 14, 2013Assignee: Tencent Technology (Shenzhen) Company LimitedInventors: Yisha Lu, Xiaxiang Lin, Jianyu Wang