公开(公告)号：US20170118170A1

公开(公告)日：2017-04-27

申请号：US15051298

申请日：2016-02-23

Applicant: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

Inventor： Ching-Yao WANG ,  Lyu-Han CHEN ,  Hung-Wei LIN

IPC: H04L29/12 ,  H04L29/08

CPC classification number: H04L61/2589 ,  H04L61/2514 ,  H04L61/2517 ,  H04L61/2575 ,  H04L67/104

Abstract: A method and a communication device for network address translation (NAT) traversal is provided. The method includes following steps. An NAT device information is exchanged between a communication device and another communication device. A relay connection is established between the communication device and the another communication device through a relay server. Whether it is feasible to establish a P2P connection between the communication device and the another communication device is determined according to the NAT device information. When it is feasible to establish the P2P connection between the communication device and the another communication device, an attempt of establishing the P2P connection between the communication device and the another communication device is made. If the P2P connection is not established successfully, an attempt of establishing the P2P connection between the communication device and the another communication device is made again.

公开(公告)号：US20220207647A1

公开(公告)日：2022-06-30

申请号：US17135818

申请日：2020-12-28

Applicant: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTE

Inventor： Hung-Wei LIN ,  Sen-Yih CHOU ,  Hsin-Cheng LIN

IPC: G06T3/40 ,  G06T5/00 ,  G06T5/20 ,  G06T7/13 ,  G06N3/08

Abstract: A training method and training system for a resolution improvement model and a boundary detection method using the resolution improvement model are provided. The training method for the resolution improvement model includes the following steps. A low-resolution image is inputted. Pixels of the low-resolution image are captured and reorganized to generate a high-resolution image according to convolutional features. The resolution of the high-resolution image is higher than that of the low-resolution image. When capturing the low-resolution image, a condition mask is used to filter off the noise content, as well as sharpen the edge. The high-resolution image is compared with a ground-truth target image to output a discrimination result. The convolutional features are updated according to the discrimination result.