Institute for Information Industry

Benq Corporation

Benq Intelligent Technology (shanghai) Co., Ltd

Kabushiki Kaisha Ubitus

Pei-Wen Huang

Jing-Fung Chen

Emery Jou

Chia-Hsiang Chang

Chin-Jui Chi

Ta-Wei Liu

Ching-Hao Mao

An-Lun Teng

Jung-Chang Kuo

Ying-Hau Wu

I-Chen Wu

Yu-Te Chen

Chiu-Chou Lin

Hsiang-An Wen

Pei Te Chen

The invention discloses a pure end-to-end deep reinforcement learning for training car racing game AI bot that uses only the velocity information extracted from screen for both training and testing phases without using any internal state from game environment, such as the car facing angle. The learned AI bot can play better than the average performance of human players. In addition, the reward function is designed to consist only the velocity value, and use Ape-X distributed training framework combined with a variant of Deep Q Network to solve the sparse training signal problem caused by the reward function of an original design. Moreover, limit learner rate method is designed that improves the training efficiency and training performance. The AI bot trained in this way can achieve performance beyond the average human level and reach a level close to professional players.

Method for training AI bot in computer game

A touch display screen includes a display panel, a touch-sensing part, a biosensor and a controller. The touch-sensing part is disposed on a light emitting side of the display panel and has a touch-sensing surface. The touch-sensing part includes a light-emitting device and a light receiver. The light-emitting device is used to provide at least one positioning light and at least one high-energy light. The light receiver is used to detect a change of the positioning light on the touch-sensing surface in response to a touch action and determine the position on which the touch action occurs on the touch-sensing surface. The biosensor is used to determine whether a living body getting close to the touch-sensing surface. The controller is used to drive the light-emitting device to selectively provide the positioning light and/or the high-energy light to the touch-sensing surface according to the determinations of the biosensor.

Touch screen

A touch control method includes steps of detecting a plurality of contact points being touched on a touch panel at a certain time; obtaining a plurality of distances between each two of the plurality of contact points; selecting a largest distance smaller than a predetermined distance among the distances; selecting the contact points falling into a first circular area formed by the two end points of the largest distance in the touch panel as a first contact point cluster and recognizing them as a first touch point; determine the first coordinate of the first touch point; selecting a second contact point cluster among the remaining contact points, by the same way, and recognizing them as a second touch point; obtaining the second coordinate of the second touch point; and performing data processing on a corresponding position of an image display area according to the first and/or second coordinate.

Touch control method and touch control system applying ihe same

A multi-screen splicing structure has a plurality of display devices. Each display device includes a display panel, a plurality of transceivers, a wireless communications module, and a control circuit. Each transceiver has a transmitting component and a receiving component. The control circuit is coupled to the display panel, the transceivers and the wireless communications module, and is used to turn on the wireless communications module when the control circuit determines that another display device is approaching based on a status of signals received by the receiving components of the transceivers to enable the wireless communications module performing wireless communications with a wireless communications module of another display device. The control circuit may also control an orientation of an image relative to the display panel according to the status of signals received by the receiving components of the transceivers of the display device.

Multi-screen splicing structure and display device having transceivers for detecting approaching of other displays

A method, an electronic device, and a user interface for on-demand detecting a malware are provided and adapted for estimating whether an application has vulnerabilities or malicious behaviors. The method includes the following steps. Firstly, evaluating a risk level and a test time of the application which has vulnerabilities or malicious behaviors. Next, detecting the application by selection of user to estimate the risk level of the application which has vulnerabilities or malicious behaviors and then correspondingly generating a detection result. Therefore, the method, the electronic device, and the user interface for on-demand detecting the malware can detect the risk level of the application which has vulnerabilities or malicious behaviors before getting virus pattern of the variant or new malware.

Method, electronic device, and user interface for on-demand detecting malware

A system and method for constructing a scene clip, and a non-statutory record medium thereof are provided. The system includes media supply equipment, a metadata server, a scene server, and an end device. The media supply equipment is used for providing media data. The metadata server is used for providing scene time information corresponding to playback scenes of the media data. A first end device acquires the media data and the scene time information, and extracts, according to capture time information input when playing the media data, at least one piece of target scene time from each piece of the scene time information. The scene server acquires the media data and the target scene time, and according to an alignment result of the target scene time and each piece of the scene time information, extracts local scene clips from the media data to form a piece of media division data.

System and method for constructing scene clip, and record medium thereof

A system and a method for summary collection and playing of scenes and a recording medium thereof are provided. The system includes media supply equipment, a scene description server, a scene server, and an end device. The media supply equipment supplies media data. The scene description server supplies scene description information of a scene segment corresponding to the media data, where each piece of scene description information records a playing content of the scene segment corresponding thereto. The scene server obtains the media data and a summary collect command supplied by the end device, and retrieves a plurality of partial scene segments from the media data according to an analysis result of the summary collect command and the playing content recorded by each piece of scene description information, so as to form media summary data and output the media summary data to the end device.

System and method for summary collection and playing of scenes and recording medium thereof

A scene segment playing system, a scene segment playing method and a recording medium thereof are provided. The system includes media supply equipment, a description generating server, a scene server, and an end device. The media supply equipment is used for supplying media data. Upon receiving the media data, the scene server supplies scene description data corresponding to the media data. The scene server acquires the media data and interval information supplied by the end device, and retrieves scene segment data from the media data according to a comparison result of the interval information and the scene description data, and outputs the scene segment data to the end device for playing.

Growing community of inventors

Taipei, Taiwan

Pei-Wen Huang