A Study on Fiducial Marker based Point Features Tracking Algorithm for Human-computer Interaction
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Recently, marker-based augmented reality technology has attracted attention in gesture based human-computer interaction due to the users' demand. However, the blur effect that occurs when the object shows fast motion in frames limits the marker detection and tracking. Conventional de-blurring techniques are not effective for real-time video without the task of extracting a specific frame from a full frame. In this paper, we proposed the algorithm which detects special marker first, and also tracks point features of markers based on optical flow method. Especially, marker tracking method extracts point features also using the FAST algorithm and calculates motion vectors by analyzing the motion of point features using Lucas-Kanade method. Also, we implemented a marker pose estimation by solving Perspective-n-Point problem. Our proposed system showed more improved results than conventional methods especially with the rate of 37% higher in the frame processing speed of marker tracking video. In addition, the marker pose estimation was graphically represented. It is expected that this research will be useful for gesture-based human-computer interaction field and mobile robot field through camera in real environment.
Abstract
Recently, marker-based augmented reality technology has attracted attention in gesture based human-computer interaction due to the users' demand. However, the blur effect that occurs when the object shows fast motion in frames limits the marker detection and tracking. Conventional de-blurring techniques are not effective for real-time video without the task of extracting a specific frame from a full frame. In this paper, we proposed the algorithm which detects special marker first, and also tracks point features of markers based on optical flow method. Especially, marker tracking method extracts point features also using the FAST algorithm and calculates motion vectors by analyzing the motion of point features using Lucas-Kanade method. Also, we implemented a marker pose estimation by solving Perspective-n-Point problem. Our proposed system showed more improved results than conventional methods especially with the rate of 37% higher in the frame processing speed of marker tracking video. In addition, the marker pose estimation was graphically represented. It is expected that this research will be useful for gesture-based human-computer interaction field and mobile robot field through camera in real environment.
참고문헌 (15)
- Jong-Seung Park, "Computer Vision Techniques for Augmented Reality Games", Journal of The Korean Society for Computer Game, (4), pp.52~57 (2004)
- MAI THANH NHAT TRUONG, "Face tracking algorithm for human-computer interactive games using color-based particle filter", 한국컴퓨터게임학회논문지, 30(1), pp.57~62 (2017), (사)한국컴퓨터게임학회[DOI]
- 이원선, "자연스런 손동작 인터렉션과 후마네트 운동을 접목한 노인성뇌질환 예방 기능성 게임기술", 한국컴퓨터게임학회논문지, 28(2), pp.37~42 (2015), (사)한국컴퓨터게임학회[DOI]
- 김준호, "모바일 게임을 위한 모바일 증강현실 기술 구현", 한국컴퓨터게임학회논문지, 26(3), pp.17~24 (2013), (사)한국컴퓨터게임학회[DOI]
- Mark Fiala, "ARTag, a fiducial marker system using digital techniques", IEEE Computer Vision and Pattern Recognition(CVPR), IEEE Computer Society Conference on, 2 (2005)
- 박명숙, "임베디드 시스템 기반의 2차원 바코드 영역 검출", 한국컴퓨터게임학회논문지, 29(1), pp.55~61 (2016), (사)한국컴퓨터게임학회[DOI]
- E. Olson, "AprilTag: A robust and flexible visual fiducial system", IEEE International Conference Robotics and Automation (ICRA), pp.3400~3407 (2011)
- 백종환, "드론과 지상로봇 간의 협업을 위한 광학흐름 기반 마커 추적방법", 정보처리학회 논문지, 7(3), pp.107~112 (2018), 한국정보처리학회[DOI]
- Bunyo Okumura, "Augmented reality based on estimation of defocusing and motion blurring from captured images", Proceedings of the 5th IEEE and ACM International Symposium on Mixed and Augmented Reality, pp.219~225 (2006), IEEE Computer Society
- Wei Xu, "2D barcod localization and motion deblurring using flutter shutter camera", IEEE Workshop on Applications of Computer Vision, pp.159~165 (2011)
- Hojin Cho, "Research Trends of Image Deblurring", The Magazine of the IEEE, 39(10), pp.25~35 (2012)
- E. Tola, "Daisy: An efficient dense descriptor applied to wide-baseline stereo", IEEE transactions on pattern analysis and machine intelligence, 32(6), pp.815~830 (2010)[DOI]
- E. Rosten, "Fusing points and lines for high performance tracking", Tenth IEEE International Conference (ICCV), 2, pp.1508~1515 (2005)
- B. D. Lucas, "An iterative image registration technique with an application to stereo vision", 7th Intl. Joint Conf. on Artificial Intelligence(IJCAI), pp.674~679 (1981)
- S. Garrido-Jurado, "Automatic generation and detection of highly reliable fiducial markers under occlusion", Pattern Recognition, 47(6), pp.2280~2292 (2014)[DOI]