An Efficient Approach for Eigen-joints-based 3D Activity Recognition
0
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
In this paper, we present an approach for activity recognition by using 3D skeleton data obtained by Kinect sensor. Primarily, we use the simplified Dynamic Time Wrapping(DTW) and calculate Euclidean geometry distance to obtain the probable activities from the trained data. Afterwards, for each activity, we define a modified activity feature descriptor using the interrelation of correlated joints in each frame. Before classification, we employ normalization to avoid nonuniformity in coordinates, and Principal Component Analysis(PCA) to deduce redundancy and decrease the dimensionality. Finally we apply Naïve-Bayes-Nearest-Neighbor(NBNN) classifier to classify multiple actions. The experimental result on benchmark dataset shows that the accuracy approximates the state-of-the-art methods.
Abstract
In this paper, we present an approach for activity recognition by using 3D skeleton data obtained by Kinect sensor. Primarily, we use the simplified Dynamic Time Wrapping(DTW) and calculate Euclidean geometry distance to obtain the probable activities from the trained data. Afterwards, for each activity, we define a modified activity feature descriptor using the interrelation of correlated joints in each frame. Before classification, we employ normalization to avoid nonuniformity in coordinates, and Principal Component Analysis(PCA) to deduce redundancy and decrease the dimensionality. Finally we apply Naïve-Bayes-Nearest-Neighbor(NBNN) classifier to classify multiple actions. The experimental result on benchmark dataset shows that the accuracy approximates the state-of-the-art methods.
참고문헌 (13)
- A. Bobick, "The recognition of human movement using temporal templates", IEEE Trans, Pattern Anal, 23(3), pp.257~267 (2002)
- Ivan Laptev, "Learning realistic human actions from movies", IEEE Conference on Computer Vision & Pattern Recognition, pp.1~8 (2008), US. IEEE Computer Society
- W. Bian, "Cross-domain human action recognition", IEEE Trans. Syst. Man Cybern, 42(2), pp.298~307 (2012)
- 나민영, "스마트 디바이스 제어를 위한 비전 기반 실시간 손 포즈 및 제스처 인식 방법", 한국차세대컴퓨팅학회 논문지, 8(4), pp.27~34 (2012), 한국차세대컴퓨팅학회
- 최재인, "특징점 기반 카메라 트래킹을 이용한 손 움직임 보정 기법", 한국차세대컴퓨팅학회 논문지, 11(2), pp.6~13 (2015), 한국차세대컴퓨팅학회
- 나민영, "실시간 손 포즈 인식을 통한 비접촉식 입력 인터페이스", 한국차세대컴퓨팅학회 논문지, 11(1), pp.25~31 (2015), 한국차세대컴퓨팅학회
- A. Klaser, "A spatiotemporal descriptor based on 3D gradients", BMVC, pp.275~284 (2008), UK, British Machine Vision Association
- Jaeyong Sung, "Human Activity Detection from RGBD Image s", AAAI workshop on Pattern, Activity and Intent Recognition, pp.47~55 (2011)
- Jaeyong Sung, "Unstructured Human Activity Detection from RGBD Images", IEEE International Conference on Robotics and Automation, 2, pp.842~849 (2012)
- L. Xia, "View invariant human action recognition using histograms of 3-D joints", IEEE Workshop CVPR Human Activity Understand, pp.20~27 (2012)
- Xia, L., "Human detection using depth information by Kinect", CVPRW, IEEE Computer Society Conference, pp.15~22 (2011)
- Reyes M, "Feature weighting in dynamic time warping for gesture recognition in depth data", IEEE, ICCV Workshops, pp.1182~1188 (2011)
- Li, W., "Action recognition based on a bag of 3d points", IEEE, CVPRW, pp.9~14 (2010)