Abstract: A method and a system for medical image interpretation are provided. A medical image is provided to a convolutional neural network model. The convolutional neural network model includes a feature extraction part, a first classifier, and N second classifiers. N feature maps are generated by using the last layer of the feature extraction part of the convolutional neural network model. N symptom interpretation results of N symptoms of a disease are obtained based on the N feature maps through the N second classifiers. A disease interpretation result of the disease is obtained based on the N feature maps through the first classifier.

Abstract: A method includes forming a first conductive feature, depositing a passivation layer on a sidewall and a top surface of the first conductive feature, etching the passivation layer to reveal the first conductive feature, and recessing a first top surface of the passivation layer to form a step. The step comprises a second top surface of the passivation layer. The method further includes forming a planarization layer on the passivation layer, and forming a second conductive feature extending into the passivation layer to contact the first conductive feature.

Abstract: The present disclosure relates to a system, a method and a computer-readable medium for rendering a streaming on a user terminal. The method includes rendering the streaming in a first mode, receiving an environment parameter of the user terminal, receiving a timing when the user terminal closes the streaming, determining a threshold value of the environment parameter based on the timing the user terminal closes the streaming, receiving an updated environment parameter of the user terminal, and rendering the streaming in a second mode if the updated environment parameter meets the threshold value. The second mode includes fewer data objects than the first mode or includes a downgraded version of a data object in the first mode for the rendering. The present disclosure can customize the rendering mode for each user and maximize the satisfaction of viewing streaming for each user.

Abstract: A method, comprising: detecting an outage of at least one functionality in a live streaming; performing an first operation toward a second user terminal; storing data of the first operation in a database of the first user terminal; and displaying an effect corresponding to the first operation during the outage. The present disclosure may store the data of operation performed by the user terminal during outage and process the operation after the outage is recovered. Therefore, the streamers and viewers may feel interested and satisfied, instead of feeling anxious, and the user experience may be enhanced.

Abstract: A wheel control mechanism includes a support base, a wheel module, a first magnetic module and a second magnetic module. The wheel module, the first magnetic module and the second magnetic module are installed on the support base. The first magnetic module generates a magnetic attractive force to attract a metal ratchet of the wheel module. By adjusting the position of the second magnetic module relative to the first magnetic module, the strength of the magnetic attractive force is changed. Consequently, the rotation of the wheel module results in a tactile feel or does not result in the tactile feel.

Abstract: The present disclosure relates to a system, a method and a computer-readable medium for rendering a streaming on a user terminal. The method includes rendering the streaming in a first mode, receiving an environment parameter of the user terminal, receiving a timing when the user terminal closes the streaming, determining a threshold value of the environment parameter based on the timing the user terminal closes the streaming, receiving an updated environment parameter of the user terminal, and rendering the streaming in a second mode if the updated environment parameter meets the threshold value. The second mode includes fewer data objects than the first mode or includes a downgraded version of a data object in the first mode for the rendering. The present disclosure can customize the rendering mode for each user and maximize the satisfaction of viewing streaming for each user.

Abstract: The subject application relates to a method and server for handling streaming data, and includes: Obtaining a request for entering the live streaming as an invisible viewer; in response to obtaining the request, starting to provide a first user terminal of a first viewer with the streaming data for the live streaming while setting information on the first viewer invisible to other viewers and a streamer; and in response to detecting a first action of the first viewer in the live streaming, setting at least a part of the information visible to at least a part of the other viewers and the streamer. According to the subject application, it is possible to encourage further communication between the streamer and the viewer, and enhance user-user interactions through the live streaming.

Abstract: The present disclosure relates to a system, a method and a computer-readable medium for rendering a streaming on a user terminal. The method includes rendering the streaming in a first mode, receiving an environment parameter of the user terminal, receiving a timing when the user terminal closes the streaming, determining a threshold value of the environment parameter based on the timing the user terminal closes the streaming, receiving an updated environment parameter of the user terminal, and rendering the streaming in a second mode if the updated environment parameter meets the threshold value. The second mode includes fewer data objects than the first mode or includes a downgraded version of a data object in the first mode for the rendering. The present disclosure can customize the rendering mode for each user and maximize the satisfaction of viewing streaming for each user.

Abstract: The present disclosure relates to a system, a method and a computer-readable medium for video processing. The method includes displaying a live video of a first user in a first region on a user terminal and displaying a video of a second user in a second region on the user terminal. A portion of the live video of the first user extends to the second region on the user terminal. The present disclosure can improve interaction during a conference call or a group call.

Abstract: An automated inspection system includes a machine-vision inspection device, a repair station and an artificial intelligence module connected to the machine-vision inspection device, and a processing platform connected to the repair station and the artificial intelligence module. The machine-vision inspection device obtains an image of a to-be-inspected object and displays the image on an operation screen of the repair station, the artificial intelligence module retrieves the image of the to-be-inspected object to perform inference for classification, the processing platform automatically overlaps result data of classification inference for the image of the to-be-inspected object on the operation screen of the repair station, and intercepts and transmits result data of manual review, which is inputted by an operator through a user interface, to the repair station.

Abstract: An automatic test method for a computer includes the following steps: reading a keyboard signal or mouse signal and a delay time in an event file in a system test directory; sending the keyboard signal or mouse signal to a to-be-tested system according to the delay time; the to-be-tested system that operates according to the keyboard signal or mouse signal sending at least one response; and verifying the response by comparing the response, which is in the form of a character string, with a character string in a correct text file in the system test directory, or sending an image acquisition signal to the to-be-tested system according to the at least one response, to acquire a screenshot, converting the screenshot into a screenshot image file, and verifying the screenshot image file corresponding to the to-be-tested system by comparing the screenshot image file with a correct screenshot image file in the system test directory.

Abstract: The present invention provides intelligent test robot system, which operation platform of intelligent robot body connected storage module for storing test script, operation platform respectively connected signal output unit and image input unit, further connected network communication module linked remote console computer. When operation platform automatically performs tests accordance test script, e.g., power switches, string inputs through keyboard, mouse cursor movements, clicks etc., it is possible to send instructions contained test script to test object via signal output unit for action controls, image input unit receives images outputted by test object and returns them to operation platform, then automatic graphic recognitions can be performed on captured images according to image recognition instructions in test script.

Abstract: An automatic test method for a computer includes the following steps: reading a keyboard signal or mouse signal and a delay time in an event file in a system test directory; sending the keyboard signal or mouse signal to a to-be-tested system according to the delay time; the to-be-tested system that operates according to the keyboard signal or mouse signal sending at least one response; and verifying the response by comparing the response, which is in the form of a character string, with a character string in a correct text file in the system test directory, or sending an image acquisition signal to the to-be-tested system according to the at least one response, to acquire a screenshot, converting the screenshot into a screenshot image file, and verifying the screenshot image file corresponding to the to-be-tested system by comparing the screenshot image file with a correct screenshot image file in the system test directory.

Abstract: A method for intellectual asset management and analysis is provided, including a method of combining intellectual asset and an expert and artificial intelligence hinting method, wherein the former includes a process of key word combination, a process of self-determined combination and a process of recombination. The process of key word combination includes acquiring a document with key words from an intellectual asset according to a key word and a pre-determined search condition to form a key word document set. The process of self-determined combination includes acquiring a self-determined index document from an intellectual asset according to a self-determined index to form a self-determined index document set. The process of recombination includes reassembling the existing key word document set or the existing self-determined index document set according to a pre-determined sifting condition. The hinting method provides a word related to the key word, acquired in a comparative related word database.