Dissertation > Excellent graduate degree dissertation topics show

Motion Capture Data Based Human Motion Synthesis

Author: LiuXingQi
Tutor: ZhaoHong；LiuWeiBin
School: Beijing Jiaotong University
Course: Computer Science
Keywords: Motion Synthesis Movement chart Movement marked Motion Analysis Movement transition Motion Capture
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 232
Quote: 0
Read: Download Dissertation

Abstract


Human motion synthesis based on motion capture data is an emphasis and difficulty in the field of virtual reality, it focuses on the reuse of existing motion capture data, the aim is to improve the availability of motion capture data, to generate a wealth of human motion sequence. Motion capture technology equipment is expensive and complicated to operate, the study of human motion synthesis based on motion capture data can well reduce human motion synthesis of the production costs, improve synthesis efficiency, has a strong research significance and application prospects. In this paper, the motion capture data around relevant key human motion synthesis technology, is designed and implemented based the movement chart human motion synthesis system. Extract calculated effective reuse of existing motion capture data, based on the characteristics of the motion data to build a movement type template, and automatic annotation of motion data, motion analysis and similarity matching calculated from the level of semantic description, in order to establish binding semantic description of the movement chart, with the movement of the transition and the interpolation operation to achieve the synthesis of human motion to generate new human motion sequences that meet user requirements. Thesis research include: research and calculation method description and characteristics of human motion characteristics. Movement sequence frame gesture on the underlying data to calculate the a typical numerical motion feature; relationship characteristics in a high-level description of the motion sequences, analysis of human movement state, to provide the semantic description of motion-based features for motion charts. Research and implementation of human motion similarity calculation method. Euclidean distance formula to calculate the similarity of the motion sequence interframe; use of dynamic time warping algorithm, completion of calculating the similarity between the different lengths motion sequence. Research and implementation of the automatic annotation of human motion data calculated. Combination of the characteristics of human body movement described by dynamic time warping and self-learning process, the establishment of a typical sports category template; motion templates and unknown motion sequence local similarity matching, automatic annotation of unknown motion sequences. Motion chart-based human motion synthesis. Sport similar calculations to build the movement chart, use the Dijkstra algorithm to carry out the the motion synthesis shortest path search motion chart; transition length calculated based on the geodesic distance, offset mapping linear calculations and spherical linear interpolation to obtain transitional motion segment; With automatic annotation of human motion data, based on the movement chart synthesis, the introduction of the semantic description of the requirements of users of synthetic sports convenient user control motion synthesis, improve the quality of motion synthesis; build a human motion synthesis system, through effective reuse existing motion capture data, generated in real time to meet user requirements, new human motion data.

Related Dissertations

  1. Theoretical Research and Simuliation Analysis of the Air Splicer on Automatic Winder,TS103.81
  2. Research of Bionic Water Strider Robot,TP242
  3. Competitive flip side swivel 360 ° motion analysis,G852
  4. TV sighting / tracking system design and implementation,TJ765.3
  5. Design and Analyze on YK8320 Cnc Gear Hobbing Machine and Study on Processing Drum Gear,TG612
  6. Based on Virtual Human automotive ergonomics evaluation of key technology research,TB18
  7. Shaker Kinematic analysis and inertia force balance studies,S225
  8. The Analysis for Mechanical Characteristics of the Spatial Crank-swing Mechanism,TH112
  9. Design and Motion Analysis of a Reconfigurable Tracked Mobile Robot,TP242.3
  10. Based on the virtual movement and motion capture natural expression of the user's body,J04
  11. The greenhouse agriculture automation work platform transmission joint research,S625
  12. Time series visualization of human motion and multi- index method,TP391.41
  13. Immersive Virtual Shooting Training System in Human-Computer Interaction Research,TP11
  14. 4R processing multi-joint wheel unloading manipulator motion analysis and research,TP241
  15. Elongated inner spiral surface NC machining process simulation study,TG659
  16. MEMS-based sensor motion capture system development and design,TP212.9
  17. Fusion depth map and 3D models of human motion capture technology research,TP391.41
  18. Learning-based human motion synthesis inverse kinematics,TP391.41
  19. Real-time virtual environment geometry, movement and behavior LOD Technology,TP391.41
  20. Effect of Different Seat Height on Lower Limb Movement of Cyclists,G872.3

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile