International Business Machines Corporation

Tinoq Inc.

Google Inc.

Beijing Xiaomi Mobile Software Co., Ltd.

Baidu Online Network Technology (beijing) Co., Ltd.

Beijing Didi Infinity Technology and Development Co., Ltd.

Boe Technology Group Co., Ltd.

Chinese Academy of Sciences

Beijing Dajia Internet Information Technology Co., Ltd.

Beijing Goldwind Science & Creation Windpower Equipment Co., Ltd.

Aol Advertising Inc.

Chinese Academy of Medical Sciences

Yang Sun

Daxiao Yu

Mir Hamid Pirahesh

Dikran S Meliksetian

Ting Yu Leung

Roberta Jo Cochrane

Jian Qiang Liu

George Lapis

Uri First

Markos Zaharioudakis

Calisto Paul Zuzarte

Zhi Jun Wang

Yan Xie

Eileen Tien Lin

Sheng Qian

Qi Cheng

Bo Xie

Lan Huang

Amit Radheshyam Somani

Zhi Hu Wang

Jing Lv

Kuanjun Peng

Nianjun Zhou

Yuan Gao

David Everett Simmen

Chuan Yang

Feng Guo

Yinglong Huang

Yuting Zhang

Tina Louise Mukai

Miao Zhang

Jing Wang

Chan-Soo Hwang

Hongliang Wang

Monica Sachiye Urata

Hans Marius Holtan

Wen Zheng

Yan Qu

Chengyu Jiang

Steven Nicholas Eliuk

Junzeng Du

Congli Song

Xiaolong Wang

Huijuan Huang

Abhishek Kolagunda

Dashan Shang

Mark A Paskin

Quan Song

Liying He

Haolin Liu

Mengfei Liu

Tianze Cui

Yisheng Chai

Chenggang Li

Shuan Rao

An image processing method, the method includes: obtaining a compressed image of a current image, where resolution of the current image is greater than that of the compressed image; cropping out at least one composition preview image based on the compressed image; and in response to selecting a target composition image from said at least one composition preview image displayed, cropping the current image based on a cropping range indicated by the target composition image.

Image processing method and device, and storage medium

An image processing method includes: determining an area to be processed according to a line feature in an image frame to be processed; building a space model according to the image frame to be processed; and embedding a special effect material to be embedded into the area to be processed according to the space model, to generate a special effect image frame.

Image processing method, electronic device, and storage medium

Implementations relate to generating and/or executing a customized interactive dialog application. The customized interactive dialog application may be generated from a state mapping tool that allows a user to generate custom states and custom transitions between the custom states. A customized configuration description is then generated based on the generated custom states and custom transitions. Further, a default configuration description is identified that includes additional or alternative states and transitions. In executing the customized interactive dialog application, dialog turns are generated based on the states and transition information, with the customized configuration description taking precedence and the default configuration description being utilized for any undefined states and/or transitions. Implementations additionally or alternatively relate to generating and/or executing a custom agent based on generated custom states and custom transitions, and a default configuration description.

User-configured and customized interactive dialog application

Various embodiments are provided for providing machine learning with noisy label data in a computing environment using one or more processors in a computing system. A label corruption probability of noisy labels may be estimated for selected data from a dataset using temporal inconsistency in a machine model prediction during a training operation in a neural network.

Intelligent learning system with noisy label data

A method for transforming hair relates to the field of image processing, and can achieve a function of transforming the hair with a more consistent and natural color with respect to that before the hair of a character is transformed. The method for transforming the hair includes: acquiring a first sample image and a second sample image; determining a first color guide map based on the first sample image and the second sample image; acquiring a first image processing model by training with the first sample image and the first color guide map as training data and the second sample image as supervision information; and transforming, based on the first image processing model, hair in an image selected by a user.

Method and apparatus for transforming hair

A magnetic pole module, an electric motor rotor and a method for manufacturing the electric motor rotor are provided. The electric motor rotor includes a rotor yoke and multiple magnetic pole modules disposed on the rotor yoke, each of the magnetic pole modules including a base plate, a housing, and a pair of magnetic pole units that are accommodated in an accommodating space formed by the base plate and the housing and have opposite polarities, and a pair of magnetic pole units in each of the magnetic pole modules are spaced apart from each other by a first distance in a circumferential direction of the rotor yoke. The electric motor rotor can give consideration to the properties, such as cogging torque and torque pulsation, of a generator, the protection of magnetic poles and the mechanical fixation of the magnetic pole.

Magnetic pole module, electric motor rotor and method for manufacturing electric motor rotor

A method for decrypting an encrypted message in a cluster may be provided. The method may include generating, by a first private key generator, one or more system parameters and a master key using a security parameter of the cluster and a depth of the maximum of a unit vector, the cluster including a first member and a second member. The method may also include generating, by the first private key generator, a private key of the first member; The method may further include generating, by a second private key generator, a private key of the second member based on the one or more system parameters, the identification vector of the first member, the private key of the first member, and an identification vector of the second member; The method may still further include decrypting the encrypted message the private key of the first member or the second member.

Systems and methods for data encryption and decryption in data transmission

A compound sensor system includes a first sensor, a second sensor, a memory that stores a module, and a processor coupled to the first sensor, the second sensor, and the memory. The first sensor is configured to detect a parameter that indicates a likelihood of having a user enter or leave a target area, and, in response, send a first command signal to the processor. The processor is configured to receive the first command signal from the first sensor and send a second command signal to the second sensor based on receiving the first command signal. The second sensor is configured to operate at a sleep mode and switch to an active mode upon receiving the second command signal, and during the active mode the second sensor is configured to determine if the user enters or leaves the target area.

Systems and methods for a compound sensor system

A method may include receiving, from a first application installed in an electronic device, a first request for accessing a first set of contents associated with a second application. The method may include obtaining a request time stamp and a content pointer associated with the first set of contents. The method may include placing the link information to a predetermined storage space in the electronic device. The method may include invoking the second application based on the content pointer and obtain an invoking time stamp. The method may include comparing a time interval between the request time stamp and the invoking time stamp to a predetermined time threshold. The method may include accessing the first set of contents in the second application according to the content pointer in response to a determination that the time interval is less than the predetermined time threshold.

Systems and methods for data processing related to an online to offline service

In response to receiving a request for visiting a first web task associated with a website, a first task node is generated at a breadcrumb root node in a tree structure. In response to sequentially receiving requests for multiple subtasks of the first web task, multiple subtask nodes of the first task node are sequentially established in the tree structure. The subtask nodes of the multiple subtask nodes of the first task node are sequentially connected to the first task node according to a sequential order of the sequentially received requests for the multiple subtasks of the first web task. The multiple subtask nodes of the first task node are processed based a policy of the first web task, which includes dynamically reconfiguring the multiple subtask nodes of the first task node in response to the user's visitation of one or more subtask nodes of the multiple subtask nodes.

Processing tree structure having breadcrumb root node

A system of user detection includes a camera configured to rotate to a plurality of positions, a memory that stores a module, and a processor coupled, locally or remotely, to the camera. The processor is configured to send a command to the camera to cause the camera to (1) rotate to a position from the plurality of positions, and (2) capture a photo or a video clip at the position. The processor is configured to receive the photo or the video clip from the camera; determine whether a user is captured in the photo or the video clip; divide the photo or the video clip into a plurality of zones; determine a zone from the plurality of zones where the user appears in the photo or the video clip; and determine whether the user is associated with a device based on information of the position and the zone.

Systems and methods for user detection and recognition

A method for recognizing the face of a user that uses a device in a facility includes receiving a photo of the user and extracting a face image of the user from the photo. The method includes receiving a full candidate face set of all registered users, a first list of registered users that are within the facility, and a second list of registered users that are using other devices in the facility. The method includes generating a reduced candidate face set from the full candidate face set by excluding at least one of (1) candidate faces corresponding to users that are not included in the first list or (2) candidate faces corresponding to users that are included in the second list, from the full candidate face set. The method also includes determining whether the face image corresponds to a candidate face of the reduced candidate face set.

Systems and methods for efficient face recognition

Embodiments of the present disclosure provide a method for controlling a smart device, a computer device and a non-transitory computer readable storage medium. The method includes performing speech recognition on a speech signal acquired by the smart device; determining whether a control instruction corresponding to the speech signal matches with a present operation scene of the smart device; and adjusting an operation state of the smart device according to the control instruction when the control instruction matches with the present operation scene.

Method for controlling smart device, computer device and storage medium

Growing community of inventors

Beijing, China

Yang Sun