Browsing by Author "Chen, Kuan-Hua"

Now showing 1 - 2 of 2

Heterogeneous Directional Sensors Self-Deployment Problem in a Bounded Monitoring Area
(2011)
Panditharathne, Charka
;
Lin, Ting-Yu
;
Chen, Kuan-Hua
;
電信工程研究所
;
Institute of Communications Engineering
Good deployment of sensors empowers the network with effective monitoring ability. Different from omnidirectional sensors, the coverage region of a directional sensor is determined by not only the sensing radius (distance), but also its sensing orientation and spread angle. Heterogeneous sensing distances and spread angles are likely to exist among directional sensors, to which we refer as heterogeneous directional sensors. In this paper, we target on a bounded monitoring area and deal with heterogeneous directional sensors equipped with locomotion and rotation facilities to enable the sensors self-deployment. Our base and optimized deployment algorithms are proposed to achieve high sensing coverage ratio in the monitored field. These algorithms leverage the concept of virtual forces (for sensors movements) and virtual boundary torques (for sensors rotations). Performance results demonstrate that our optimized deployment mechanism is capable of providing desirable surveillance level, while consuming moderate moving and rotating energy.
REAL-TIME UPPER BODY POSE ESTIMATION FROM DEPTH IMAGES
(2015)
Tsai, Ming-Han
;
Chen, Kuan-Hua
;
Lin, I-Chen
;
交大名義發表
;
National Chiao Tung University
Estimating upper body poses from a sequence of depth images is a challenging problem. Lately, the state-of-art work adopted a randomized forest method to label human parts in real time. However, it requires enormous training data to obtain favorable results. In this paper, we propose using a novel two-stage method to estimate the probability maps of upper body parts of users. These maps are then used to evaluate the region fitness of body parts for pose recovery. Experiments show that the proposed method can obtain satisfactory outcome in real time and it requires a moderate size of training data.