Hong Kong Applied Science and Technology Research Institute Company, Limited

The Chinese University of Hong Kong

Yan Wang

Soung Liew

Meixia Tao

Piu Bill Wong

Ross David Murch

Yang Liu

Kin Nang Lau

Zhi Bin Lei

Hang Ching Jason Leung

Hailiang Li

Douglas Ronald George

Vincent Lau

Man Yau Chiu

Soung Chang Liew

Cheong Yui Wong

David Chi Cheong Wan

Kent Chen

Wai Kit Law

Yi Gong

Jian Shu Hu

Mario Joa-Ng

Tsz Ming Ip

Methods and systems which provide super-resolution synthesis based on weighted results from a random forest classifier are described. Embodiments apply a trained random forest classifier to low-resolution patches generated from the low-resolution input image to classify the low-resolution input patches. As each low-resolution patch is fed into the random forest classifier, each decision tree in the random forest classifier 'votes' for a particular class for each of the low-resolution patches. Each class is associated with a projection matrix. The projection matrices output by the decision trees are combined by a weighted average to calculate an overall projection matrix corresponding to the random forest classifier output, which is used to calculate a high-resolution patch for each low-resolution patch. The high-resolution patches are combined to generate a synthesized high-resolution image corresponding to the low-resolution input image.

Systems and methods for super-resolution synthesis based on weighted results from a random forest classifier

Using molecular docking screening, a few therapeutical compounds including trovafloxacin and levofloxacin have been first identified as potential inhibitors of fatty acid binding protein 4 (FABP4). The present invention provides a novel use of levofloxacin as an inhibitor of FABP4 and a drug for the treatment of metabolic diseases or a cardiovascular diseases. The present invention further provides a method of using levofloxacin to inhibit FABP4, or to inhibit adipolysis in adipocytes.

Discovery of FDA-approved drugs as inhibitors of fatty acid binding protein 4 using molecular docking screening

A method of determining an edge of an object on a digital image sequence comprising the step of determining a first gradient direction profile of a first image in the digital image sequence; determining a second gradient direction profile of a second image in the digital image sequence; computing a differential profile based on the first gradient direction profile and the second gradient direction profile; and determining the edge of the object based on the differential profile wherein the differential profile registers gradient magnitudes of the second gradient direction profile and angular differences between the first gradient direction profile and the second gradient direction profile. A system thereof is also disclosed.

Method of detecting edge under non-uniform lighting background

Disclosed are systems and methods which provide high bandwidth data communication with respect to a large coverage area using smart antenna and/or directional antenna (referred to herein as multi-beam antenna) technology. Circuitry may be provided at a WLAN AP to provide selection of particular antenna beams used in the downlink and/or uplink, control of multicast transmission, control of unicast transmission, and to provide antenna pattern shaping techniques. Embodiments implement multi-beam antenna technology with little or no hardware modifications to AP circuitry. Other embodiments implement multi-beam technology using radio front-end and/or radio hardware modifications to AP circuitry. Various diversity techniques may be implemented, such as selection diversity, maximum ratio combining, and equal gain combing. To provide desired antenna pattern shaping, phase offsets with respect to a signal as transmitted in each antenna beam may be employed.

WLAN access point with extended coverage area

Disclosed are systems and methods which provide a busy tone or signal for use in communication collision avoidance in a wireless network, such as a WLAN, a cellular network, or other point-to-multipoint wireless network. An intelligent busy tone generator, upon the proper conditions, transmits a busy tone or signal to one or more wireless device which may otherwise be hidden from particular communications. According to embodiments the busy tone mimics a carrier sequence of an actual transmission to facilitate the wireless devices recognizing the channel as busy. Because the original uplink packet may be interfered by transmission of a busy tone carrier-sense sequence, embodiments are provided with interference cancellation capability. Interference cancellation circuitry of embodiments may operate in a plurality of modes, including a training mode (including an online training mode, an offline training mode, and a hybrid or combined mode) and an operating mode.

Busy tone for wireless networks

Disclosed are systems and methods which provide for location positioning in wireless networks using techniques which are adapted to provide reliable location determinations even in complex topological environments. Embodiments utilize multiple antenna patterns, such as may be provided using phased array antennas, and implement location positioning techniques which do not require alteration of remote stations in providing location positioning. Various techniques for determining location may be implemented, including a channel model independent approach, a channel model based approach, or combinations thereof. A channel model independent approach used in providing location positioning may compare receive signal strength differences to an antenna gain difference table to determine an angle in the azimuth that a remote station is located. A channel model based approach used in providing location positioning may compare receive signal strength measurements to a database of signal strengths created using a wireless channel model.

Location positioning in wireless networks

Growing community of inventors

Hong Kong, China

Yan Wang