Dissertation > Excellent graduate degree dissertation topics show

The Research of Some Key Technologies in the 3D HCI System

Author: LiYang
Tutor: PanZhiGeng
School: Zhejiang University
Course: Applied Computer Technology
Keywords: Augmented Reality Registration Natural Interaction Bare-hand Tracking 3D Reconstruction
CLC: TP11
Type: Master's thesis
Year: 2011
Downloads: 241
Quote: 0
Read: Download Dissertation

Abstract


Being probably the most natural HCI method,3D natural interaction has attracted more and more attention in recent years. Compared with traditional mouse and keyboard devices, the 3D natural interaction makes the interaction system more immersive and easy to interact.We mainly focus on the two key technologies of the 3D HCI system:Planar target based AR registration and natural hand gesture interaction.For the Planar target based AR registration, a robust natural feature based parallel registration method is proposed. The method uses two threads for registration on a dual-core computer, one thread using KLT for tracking, while the other using Surf to reset the accumulated-errors. Delayed homograph is proposed to solve the problem of thread synchronization. At the same time, the idea of key-frame increased the accuracy and stability of the registration. The experiment results proved that the multi-thread method solved the problem of accumulated-errors, and it is a robust feature based real-time registration method.For the natural hand gesture interaction, two interaction situation are considered, in the situation of single camera interaction, we proposed an easy-to-use and inexpensive approach to track the hands accurately. Outstretched hand is detected by contour & curvature based detection techniques to initialize the tracking region. Robust multi-cue hand tracking is then achieved by velocity-weighted features and color cue. Experiments show that the proposed multi-cue hand tracking approach achieves continuous real-time results even for the situation of cluttered background. In the situation of two camera 3D hand gesture interaction, instead of reconstructing the whole hand structure, we extract some robust features from the 2D bare-hand images, and then reconstruct these features for 3D interaction. These features are concluded as Point, Line, Planar and Simple-Gesture interactions. We applied these features to several applications in AR systems and computer games for interaction. Experiments show that the accuracy and robustness can meet the requirements of the desktop

Related Dissertations

  1. Application of a Three Dimensional Digital Model for Hallux Valgus Osteotomy,R687.3
  2. Pose-Estimation Based on Projective Reconstruction for Augmented Reality,TP391.41
  3. Study of the Key Technology of Augmented Reality,TP391.9
  4. Camera Sync and particle trajectories to obtain three-dimensional,TP391.41
  5. The Research of Human Motion Analysis System Based on Video,TP391.41
  6. Augmented reality technology research registration,TP391.41
  7. 3D Model Reconstruction Based on 2D Images,TP391.41
  8. Research on Three-Dimensional Visualization of Medical Image,TP391.41
  9. Research for Mix Expression on the Scene Indoor Using Polygon and Grid,TP391.41
  10. Research on Related Technologies for 3D Reconstruction from Uncalibrated Image Sequences,TP391.41
  11. 3D Measurement Technology Based on Shape from Silhouette,TP391.41
  12. Large Size and High Temperature Forging Binocular Stereo Vision Measurement Method,TP391.41
  13. Seed Point Acquisition of 3D Facial Data Based on Binocular Stereo Vision,TP391.41
  14. Research on the Technique of 3D Reconstruction and Measurement of Vertebra Based on CT Image,TP391.41
  15. A Study on the Application of Compression and 3-D Reconstruction for Images of Multifocal Multiphoton Microscopy,TP391.41
  16. Utility of 256 Slice Intelligent CT in the Diagnosis of Cricoarytenoid Dislocation,R816.96
  17. The Study of Sex Determination of Sacral Basis in CT 3D Reconstruction,D919.4
  18. Research and Implementation of 3D Rconstruction from Uncalibrated Digital Images,TP391.41
  19. Strain Measurement of Sheet Metal Parts Based on Multiple Grid Images Taken by One Single Hand-held Digital Camera,TP391.41
  20. Research on STEP-based Manufacturing Feature Recognition and 3D Reconstruction Technology,TP391.7
  21. Binocular stereo vision based on three-dimensional face modeling technology research,TP391.41

CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Automation systems theory
© 2012 www.DissertationTopic.Net  Mobile