Search

Object detection in a road scene has received a significant attention from research fields of developing autonomous vehicle and automatic road monitoring systems. However, object occlusion problems frequently occur in generic road scenes. Due to such occlusion problems, previous object detection methods have limitations of not being able to detect objects accurately. In this paper, we propose a novel object detection network which is robust in occlusions. For effective object detection even with occlusion, the proposed network mainly consists of two parts; 1) Object detection framework, 2) Multiple object bounding box (OBB)-Critic network for predicting a BB map which estimates both ob-j ect region and occlusion region. Comprehensive experimental results on a KITTI Vision Benchmark Suite dataset showed that the proposed object detection network outperformed the state-of-the-art methods.

People use more and more electronic devices at home, in their car or in the office and with the proliferation of smart connected devices. The internet of things (IoT) is a vision of interconnected everyday objects and self-configuration wireless networks of sensors. In this paper, we describe a new concept of the social IoT as a pervasive interaction paradigm for the information sharing in the Web of devices. And we propose the architecture of the system for information sharing in the environment of IoT.

Recently, many mobile techniques such as sensor networks or various types of mobile devices make it possible to provide smart services at any time, and anywhere. In despite of these remarkable advances of techniques, there are few personalized mobile recommendation services which fully consider user's current situation. Proposed recommendation algorithm efficiently defines user's current situation with situational data captured from various smartphone sensors. Also, the algorithm uses user's social network for efficiently filtering valuable items which are considered as authorities. To verify the usefulness of proposed technique, we implement a prototype of the personalized music recommendation service in which proposed recommendation technique is applied. Additionally, through the demonstration of implemented prototype, we investigate the effect of incorporating smartphone sensor data and social data to collaborative filtering algorithms.

There are many attempts to optimize information retrieval using various social concepts such as social annotations and social book marking. This paper explores the convergence of the information retrieval technique, the social network and the context-aware computing in order to improve information retrieval services. We propose a new ranking algorithm for the context-aware information retrieval. To do this, we get ideas from the social network, previous graph theory and the context-aware computing. Therefore, we investigate the problem of the information retrieval system in ubiquitous computing environment.