Electronics and Telecommunications Research Institute

Ji-Young Kwak

Seon-Tae Kim

Chang-Sik Cho

Kyung-Hee Lee

Jae-Bok Park

Ik-Soo Shin

Jin-Wuk Seok

Pyeong-Soo Mah

Hyun-Woo Cho

Hong-Soog Kim

Mi-Sun Yu

Disclosed herein are a neural network model deployment method and apparatus for providing a deep learning service. The neural network model deployment method may include providing a specification wizard to a user, searching for and training a neural network based on a user requirement specification that is input through the specification wizard, generating a neural network template code based on the user requirement specification and the trained neural network, converting the trained neural network into a deployment neural network that is usable in a target device based on the user requirement specification, and deploying the deployment neural network to the target device.

Neural network model deployment method and apparatus for providing deep learning service

An apparatus and method for generating a neural network executable image are disclosed. The apparatus receives user requirements including a default neural network model and training result data for generating a neural network executable image required by a user, checks whether the default neural network model included in the user requirements is capable of being supported in a target system in which the neural network executable image is to be installed, converts the default neural network model into a neural network model executable in the target system, converts the training result data by reconfiguring the data format set of the training result data, and generates a neural network executable image by combining the converted neural network model and the converted training result data.

Apparatus and method for generating executable image of neural network

Disclosed herein are an apparatus and method for developing a neural network application. The apparatus includes one or more processors and executable memory for storing at least one program executed by the one or more processors. The at least one program receives a target specification and an application specification including user requirements, searches for a neural network model corresponding to the target specification and the application specification in a database, builds an inference engine for performing a neural network operation used by the neural network model, and generates a target image for executing the neural network model to be suitable for a target device using the inference engine.

Apparatus and method for developing neural network applications

Disclosed herein is a method for setting up active networking of smart devices for providing a converged service. In the method of setting up active networking of smart devices for providing a converged service, each of a plurality of smart objects broadcasts location information thereof. A smart object neighboring a largest number of smart objects is selected as a smart zone manager from among the plurality of smart objects. The selected smart zone manager broadcasts smart zone information to neighbor smart objects, and then forms a smart zone.

Apparatus and method for setting up active networking of smart devices for providing converged service

Disclosed herein is an ad-hoc multimedia group communication terminal which is robust to packet loss and includes a multimedia transmitter. The media transmitter includes a reception status monitoring module, an error recovery level/traffic adjustment module, a bit stream division and error recovery data generation module, and a media data packet generation module. The reception status monitoring module recognizes packet reception status information relevant to the reception side terminals based on control packets from each of the reception side terminals. The error recovery level/traffic adjustment module adjusts the error recovery level and the amount of traffic of the media data packets. The bit stream division and error recovery data generation module generates error recovery data to be transmitted to the reception side terminals based on the error recovery level. The media data packet generation module generates the media data packets and error recovery data packets.

Growing community of inventors

Gwangju, South Korea

Ji-Young Kwak