Dissertation > Excellent graduate degree dissertation topics show

The Research of Three-dimensional Motion Retrieval Based on SOM Feature

Author: LiYuMei
Tutor: WeiXiaoPeng；ZhangQiang
School: Dalian University
Course: Applied Computer Technology
Keywords: Human Motion Capture Database SOM PCA Weighted Mahalanobis distance 3D Motion Retrieval
CLC: TP391.41
Type: Master's thesis
Year: 2011
Downloads: 42
Quote: 0
Read: Download Dissertation

Abstract


The step-by-step development of motion capture technology and equipment technology advances greatly promoted the formation of a large number of three-dimensional human motion capture data , its field of application is also widely expanded to computer animation , film special effects . Therefore, the graphics field and animation applications research focus turned to the study of human motion capture data . However, prior to the various types of processing in the motion capture data , must be able to start with the existing human motion capture database to retrieve the required movement , and to quickly and accurately . Therefore, how to make effective use of computer technology , from motion capture database quickly and accurately retrieve the needs of a variety of sports is an urgent problem . This article is in the use of self-organizing map neural network (SOM) of the topological characteristics of human motion capture data feature maps based on proposed based on SOM characteristics and principal component analysis (PCA) index is a combination of three-dimensional motion retrieval based on SOM 3D motion retrieval of two algorithms combining the characteristics and weighted Mahalanobis distance . In order to simplify the process of movement retrieved using SOM feature map to achieve feature extraction and data dimension reduction combined with ordinary SOM topological characteristics must be to enhance the process in order to carry out the feature extraction . Largest eigenvectors the SOM mapping each movement to the surface of the feature after a train of thought is to use the PCA algorithm to extract features curved indexing mechanism to speed up the retrieval rate ; One idea is extracted by PCA on the basis of the feature surface principal component , then the use of the principal component of the contribution rate to determine the weighted Mahalanobis distance weights , and finally calculate the weighted Mahalanobis distance similarity comparison . The two algorithms were simulated and compared , and the experimental results demonstrate the effectiveness of the two algorithms .

Related Dissertations

  1. The Research of the Fault Diagnoses Algorithm for the Liquid Rocket Engine Testing Bed Based on PCA-SVM,V433.9
  2. Research on Recognition of Fabric Weave Pattern Based on Space-Frequency Domain,TS101.923
  3. The Research on Evaluation of Living Status Systems of Expressway Relocated People,D523
  4. GC-MS Analysis of Endogenous Metabolites in Plasma of Police Dog after Aflatoxicosis,S858.292
  5. A Preliminary Study on Bare Soil Extraction and Its Moisture Retrieval from High Resolution SAR Data,S152.7
  6. Bag-of-Visual Words for Image Categorization,TP391.41
  7. Application of Functional MR Imaging in Diagnose for Prostate Cancer,R737.25
  8. Image Restoration Method Research and Its Application Based on Probability PCA,TP391.41
  9. Design and Implementation of the Process Monitoring and Fault Diagnosis for Injection Molding,TQ320.5
  10. Research on Image Super-resolution Reconstruction Based on Non-local Similarity,TP391.41
  11. The Research on Neural Network with Quadratic Denominator Cubic Rational Spline Function Weight and Its Application,TP183
  12. Multi-step-ahead Stock Price Index Forecasting Based on Hybrid Models,F224
  13. Hash based on structured sparse spectrum image indexing algorithm,TP391.41
  14. Path Planning of Robot Systems,TP242
  15. Based on spectral regularization of linear dimensionality reduction methods,TP391.41
  16. Context of Chinese 3G mobile customer relationship and satisfaction evaluation study,F626;F274
  17. Based on PCA and SVM automotive coating line electromechanical equipment intelligent diagnosis,TH165.3
  18. Three-dimensional seismic signal density point number combination technology research,TP301.6
  19. Research on the Surface Enhanced Raman Spectroscopy of Saliva for the Diagnosis of Lung Cancer,R734.2
  20. Remote Sensing Image Fusion Technology,P237
  21. Study on the Construction Enterprises Admittance System from Governance Perspective,F426.92

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