Jun-Ho Choi 교수

Abstract The research looks at the structure of the Internet backbone and air transport networks between 82 cities in 2002, using Internet backbone bandwidth and air passenger traffic data. Centrality measures on individual city's hierarchy in the Internet and in the air traffic networks were significantly correlated, with London in the most dominant position in both networks. A quadratic assignment procedure (QAP) showed a structural equivalence between two systems. The division and membership

This study investigated cross-media credibility perception with respect to news coverage about the Iraq War. In an environment of political partisanship, perceptions of media credibility were likely affected by the audience's political position on the war. Based on hostile media effect theory, a set of hypotheses was proposed to investigate whether the minority opinion group, war opponents, evaluated the Internet as a more credible medium than did neutrals or supporters. An online survey was con

This study examined the global structure of intercultural communication on a computer-mediated communication network. Extracted from a total of 232,479 discussion messages, a matrix of crossposted messages among 133 online newsgroups over a year on the Usenet was analyzed to investigate structural patterns of communication flow. This research found, unlike earlier research, that a simple structure of core-periphery relations does not fit the pattern of cross-cultural postings in Usenet discussio

Human activity recognition using multimodal sensors is widely studied in recent days. In this paper, we propose an end-to-end deep learning model for activity recognition, which fuses features of multiple modalities based on their confidence scores that are automatically determined. The confidence scores efficiently regulate the level of contribution of each sensor. We conduct an experiment on the latest activity recognition dataset. The results confirm that our model outperforms existing method

Successful meetings create a safe environment for contribution; one that attendees feel engaged in and part of. Previous research has shown that meetings success depends not only on execution, but also on whether attendees feel psychologically safe. While this aspect is, to a great extent, partly observable through certain body cues during in-person meetings, they are often overlooked in virtual ones. To partly fix that, we developed "Kairos"-a system for multi-modal monitoring of virtual meetin

Human activity recognition using multiple sensors is a challenging but promising task in recent decades. In this paper, we propose a deep multimodal fusion model for activity recognition based on the recently proposed feature fusion architecture named EmbraceNet. Our model processes each sensor data independently, combines the features with the EmbraceNet architecture, and post-processes the fused feature to predict the activity. In addition, we propose additional processes to boost the performa

In these days, a large number of videos is taken by various kinds of handheld devices, but many of them have poor aesthetic quality. In this paper, we present an automated video editing system that uses the shot length, camera motion, and color distribution as key aesthetic features. Given an amateur video, our system computes the original unrefined camera motion as homography and tries to remove some unreliable frames, which consequently splits the video into several shots. It then applies enha

Single-image super-resolution aims to generate a high-resolution version of a low-resolution image, which serves as an essential component in many image processing applications. This paper investigates the robustness of deep learning-based super-resolution methods against adversarial attacks, which can significantly deteriorate the super-resolved images without noticeable distortion in the attacked low-resolution images. It is demonstrated that state-of-the-art deep super-resolution methods are

Recently, the vulnerability of deep image classification models to adversarial attacks has been investigated. However, such an issue has not been thoroughly studied for image-to-image tasks that take an input image and generate an output image (e.g., colorization, denoising, deblurring, etc.) This paper presents comprehensive investigations into the vulnerability of deep image-to-image models to adversarial attacks. For five popular image-to-image tasks, 16 deep models are analyzed from various