虚拟现实飞行模拟训练中的视觉交互技术研究
收稿日期: 2013-01-11
修回日期: 2013-05-22
网络出版日期: 2013-06-17
基金资助
国家自然科学基金(51305255,61039002);上海市自然科学基金(13ZR1455900)
Research of Visual Interaction for Virtual Reality Flight Training
Received date: 2013-01-11
Revised date: 2013-05-22
Online published: 2013-06-17
Supported by
National Natural Science Foundation of China (51305255,61039002);Natural Science Foundation of Shanghai City (13ZR1455900)
为实现半虚拟现实座舱环境中的人机自然交互,使用基于表观的视觉手姿态估计方法。针对基于表观方法在大容量高维度手势特征检索过程中所存在的时间效率低、内存消耗大等问题,提出了引入多探测原理和近邻特征表的改进局部敏感哈希(LSH)索引,并提出了索引性能预测模型和基于性能预测模型的索引参数优化方法,提高了索引方法的检索性能。实验结果表明,预测模型能反映实际的索引性能,使用参数优化后的改进LSH索引进行10近邻特征检索,可保证索引召回率基本不变,而使在线实际总耗时减少41.9%。将改进LSH索引应用于视觉手姿态估计,可实现虚拟手可视化,再现用户真实手的各种动作和状态。
周来 , 郑丹力 , 顾宏斌 , 王得宝 . 虚拟现实飞行模拟训练中的视觉交互技术研究[J]. 航空学报, 2013 , 34(10) : 2391 -2401 . DOI: 10.7527/S1000-6893.2013.0290
Appearance-based hand pose estimation which relies on computer vision techniques is adopted to realize natural interaction in a semi-virtual reality cockpit. To cope with the low efficiency and high memory consumption in large capacity and high-dimension feature indexing, an improved locality sensitive Hashing (LSH) method is proposed in this paper which combines the multi-probe principle with the nearest-neighbor table. Moreover, a forecast model which predicts indexing performance and a parameter optimization method are used to achieve better indexing performance. Experimental results show that the forecast model is appropriate for practical indexing performances and the time consumption is reduced by 41.9% at the cost of a slight recall rate drop. In summary, the application of the improved LSH to hand pose estimations able to upgrade virtual hand visualization and hand posture reconstruction in a semi-virtual reality cockpit.
[1] Gu H B, Wu D S, Liu H. Development of a novel low-cost flight simulator for pilot training. World Academy of Science, Engineering and Technology, 2009, 60: 685-689.
[2] Han Y. A low-cost visual motion data glove as an input device to interpret human hand gestures. IEEE Transactions on Consumer Electronics, 2010, 56(2): 501-509.
[3] Wang B, Dai S L. Experience based grasping gesture synthesis and data glove calibration. Journal of Beijing University of Aeronautics and Astronautics, 2010, 36(9):1084-1088. (in Chinese) 王滨, 戴树岭. 基于经验的抓取姿态构造及数据手套校正. 北京航空航天大学学报, 2010, 36(9): 1084-1088.
[4] Zhou L, Gu H B, Sun J, et al. Hand pose estimation based on maximum a posteriori framework in semi-virtual reality cockpit. Acta Aeronautica et Astronautica Sinica, 2011, 32(7): 1252-1259. (in Chinese) 周来, 顾宏斌, 孙瑾, 等. 半虚拟现实座舱中基于最大后验概率框架的手姿态估计. 航空学报, 2011, 32(7): 1252-1259.
[5] Causo A, Matsuo M, Ueda E, et al. Hand pose estimation using voxel-based individualized hand model. IEEE/ASME International Conference on Advanced Intelligent Mechatronics, 2009: 451-456.
[6] De La Gorce M, Fleet D J, Paragios N. Model-based 3D hand pose estimation from monocular video. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, 33(9): 1793-1805.
[7] Athitsos V, Sclaroff S. An appearance-based framework for 3D hand shape classification and camera viewpoint estimation. IEEE International Conference on Automatic Face and Gesture Recognition, 2002: 40-45.
[8] Nolker C, Ritter H. Visual recognition of continuous hand postures. IEEE Transactions on Neural Networks, 2002, 13(4): 983-994.
[9] Shih B Y, Lo T W, Chen C Y. The research of quad-tree search algorithms for anti-collision in radio frequency identification systems. Scientific Research and Essays, 2011, 6(25): 5342-5350.
[10] Zhang X B, Cheng L L, Zhu Q M. Improvement of filtering algorithm for RFID middleware using KDB-tree query index. Journal of Software, 2011, 6(12): 2521-2527.
[11] Kao B, Lee S D, Lee F KF, et al. Clustering uncertain data using voronoi diagrams and R-tree index. IEEE Transactions on Knowledge and Data Engineering, 2010, 22(9): 1219-1233.
[12] Venkateswaran J, Subramanya S R. Schemes for SR-tree packing. Proceedings of the ACM Symposium on Applied Computing, 2006: 73-77.
[13] Berchtold S, Bohm C, Kriegal H P. The pyramid-technique: towards breaking the curse of dimensionality. ACM Special Interest Group on Management of Data Record, 1998, 27(2): 142-153.
[14] Lavergne P, Patilea V. Breaking the curse of dimensionality in nonparametric testing. Journal of Econometrics, 2008, 143(1): 103-122.
[15] Cui J T, Tang J X, Jiang L. Efficient multi-resolution histogram matching for bag-of-features. IEEE International Conference on Image and Graphics. Washington D.C.: IEEE Computer Society, 2011: 406-411.
[16] Paliwal A, Jayaraman U, Gupta P. A score based indexing scheme for palmprint databases. 2010 17th IEEE International Conference on Image Processing, 2010: 2377-2380.
[17] Indyk P, Motwani R. Approximate nearest neighbors: towards removing the curse of dimensionality. Proceedings of the 13th Annual ACM Symposium on Theory of Computing, 1998: 604-613.
[18] Gionis A, Indyk P, Motwani R. Similarity search in high dimensions via hashing. International Conference on Very Large Data Bases, 1999: 518-529.
[19] Datar M, Immorlica N, Indyk P, et al. Locality-sensitive hashing scheme based on p-stable distributions. Proceeding of the 20th Annual Symposium on Computational Geometry, 2004: 253-262.
[20] Lv Q, Josephson W, Wang Z, et al. Multi-probe LSH: efficient indexing for high-dimensional similarity search. Proceedings of the 33rd International Conference on Very Large Data Bases. 2007: 950-961.
[21] Dong W, Wang Z, Josephson W, et al. Modeling LSH for performance tuning. Proceedings of the 17th ACM Conference on Information and Knowledge Management, 2008: 669-678.
[22] Zhou L, Gu H B, Tang Y. Hand pose estimation in head-mounted display virtual environments. Journal of Computer-Aided Design & Computer Graphics, 2011, 23(7): 1179-1185. (in Chinese) 周来, 顾宏斌, 汤勇. 头盔式虚拟环境中的人手姿态参数估计方法. 计算机辅助设计与图形学学报, 2011, 23(7): 1179-1185.
[23] Andoni A, Indyk P. Near-optimal hashing algorithms for approximate nearest neighbor in high dimensions. IEEE Symposium on Foundations of Computer Science, 2006: 459-468.
/
| 〈 |
|
〉 |
CJA