Sungkyunkwan University

Hong Uk Woo

Ik Jun Yeom

Won Je Choi

Min Jong Yoo

Gwang Pyo Yoo

Young Hoon Kim

Sung Hyun Choi

Chan Gue Jung

Su Hwan Kim

Min Jae Park

Seung Hwan Jeong

Young Seok Lee

Yun Ho Kim

An augmented reality (AR) share system includes a device configured to select a predetermined number of specific frames, among a plurality of frames used for a registration process of a cloud anchor; and a server configured to receive the predetermined number of specific frames, and generate an auxiliary plane anchor defining a position of a virtual object based on the predetermined number of specific frames.

Augmented reality (AR) share system

According to an exemplary embodiment of the present invention, a reinforcement learning method using a task decomposition inference model in a time-variant environment includes selecting a plurality of paired transitions having a time-invariant common characteristic and a time-variant different environmental characteristic from a dataset including a plurality of transition data, based on a cycle generative adversarial network (GAN), training an auto encoder to embed each of the time-variant part and the time-invariant part with respect to the plurality of paired transitions into a latent space, and performing reinforcement learning on a transition corresponding to data collected in the time-variant environment, using the trained auto encoder.

Reinforcement learning method and apparatus using task decomposition

A reinforcement learning-based sensor data management system includes a processor configured to: manage virtualized objects that correspond to sensors included in a sensor network to update data received from each sensor and queries representing a data quality requested by an application; calculate an abstracted action that abstracts a size of an action space of the sensor network based on present state information of the virtualized objects and the queries; calculate scores for virtualized objects based on position relationships between the calculated abstracted action the virtualized objects; and assign priorities to the virtualized objects based on the calculated scores to update data received from the sensors to the virtualized objects according to the priorities.

Reinforcement learning-based sensor data management method and system

The present disclosure is an apparatus and a method for providing a precise motion estimation learning model including, a database unit which stores a standard dataset labeled according to a first number of key points, an animation dataset labeled according to a second number of key points which is larger than the first number, and a photorealistic dataset having the second number of key points, a standard learning unit which learns the standard dataset for motion estimation to generate a standard learning model, an animation learning unit which retrains the animation dataset based on a weight of the standard learning model to generate an animation learning model, and a motion estimation learning unit which trains the photorealistic dataset based on the weight of the animation learning model to finely tune to generate a precise motion estimation learning model.

Apparatus and methods for providing precise motion estimation learning model

The disclosure relates to a method and device for inputting and outputting packets inside a GPU based on a commodity Ethernet device. According to embodiments of the disclosure, a method for commodity Ethernet device-based graphic processing unit (GPU) internal packet input/output performed by a GPU internal packet input/output device comprises being allocated an available packet buffer from a memory pool inside a GPU, after packets received from a network interface controller (NIC) are directly transferred to the allocated packet buffer, processing the directly transferred packets through a reception (Rx) kernel, transmitting a transmission packet to a network through the NIC according to an operation of a transmission (Tx) kernel, and returning the allocated packet buffer.

GPU-native packet I/O method and apparatus for GPU application on commodity ethernet

The present invention provides a graph embedding-based virtual network mapping method including inputting a virtual network and a substrate network corresponding to a virtual network mapping scenario, in an optimized mapping mode, generating an embedding value for every substrate node by applying a graph convolution network (GCN), by a network encoder, to embed the virtual network, determining whether a difference between an embedding value for every substrate node and an embedding value for every previous substrate node which is previously embedded exceeds a set threshold, by a network decoder, and mapping an allocation node of the virtual network to a mapping node of a previous substrate network which is previously mapped according to an embedding value for every previous substrate node when the difference value does not exceed the threshold value, by the network decoder.

Growing community of inventors

Suwon-si, South Korea

Hong Uk Woo