http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
Recognizing Facial Expression for Man-machine Interaction
Wataru Hirata,Joo Kooi Tan,Hyongseop Kim,Seiji Ishikawa 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
In this paper, we propose a positive face expression recognition method based on the image around the mouth. We also estimate the direction of the face of a person to examine whether or not he/she has interest toward this side. In the proposed method, estimation of the position and the direction of a face is realized by using particle filter and FAST operator. The image around the mouth is detected based on the estimated face position. The feature is calculated by Gabor Filter and the facial expression is recognized by support vector machine. In the recognition of positive expression like smile or laugh, we confirmed the effectiveness of the proposed method.
Entire 3-D Modeling of an Object by Surround Cameras
Toshimasa Sone,Joo Kooi Tan,Hyongseop Kim,Seiji Ishikawa 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
This paper proposes a novel 3-D modeling technique of an object. The existent 3-D modeling techniques such as stereo vision or the factorization recover 3-D shapes of part of an object observed commonly from multiple orientations. This inevitably needs registration among recovered partial shapes in order to obtain an entire 3-D model of the object. The proposed technique recovers entire shape of an object without registration by the employment of the cameras that surround the object. The cameras are calibrated using the captured images. Experimental results show satisfactory performance of the technique.
Robust Human Motion Recognition Employing Adaptive Database Structure
S. M. Ashik Eftakhar,Joo Kooi Tan,Hyongseop Kim,Seiji Ishikawa(편집자) 제어로봇시스템학회 2009 제어로봇시스템학회 국제학술대회 논문집 Vol.2009 No.8
Development of a robust human motion recognition system concerns with the fact that it adapts with the complexities of recognition. As the determination of the direction of motion is a significant cue, this paper employs the directional vectors in motion analysis that claims the enhancement of robustness over earlier methods. Multi-directional distinct motions are represented and compressed with the motion flow detection and compression technique, and prominent features are extracted. The extracted features are stored within an adaptive database. Finally, the evaluation of the proposed system is performed by recognizing different motions. The results guarantee the robustness of the system.