Dissertation > Excellent graduate degree dissertation topics show

3D Facial Reconstruction Based on Single Frontal Photo

Author: ChenZuo
Tutor: WanLin
School: Huazhong University of Science and Technology
Course: Software Engineering
Keywords: Realistic 3D face model Single photo Recognition algorithm CEAIP Facial feature extraction grid model Surface smoothing algorithm
CLC: TP391.41
Type: Master's thesis
Year: 2010
Downloads: 93
Quote: 0
Read: Download Dissertation

Abstract


People face daily human emotional expression, the most important and the most direct exchange carrier, a realistic three-dimensional human face has broad application prospects through computer-generated, computer graphics, computer vision, artificial intelligence and other fields in recent years, the most One of the challenging problems. With the development of computer graphics, realistic 3D face reconstruction of the increasingly wide range of applications in the field of film and television, video games and virtual environments, para-medical, criminal identification. Realistic 3D face modeling research is increasingly great theoretical significance and application value. Automatic 3D face modeling technology based on a single frontal face photo is a simple face modeling method, the method simply input a neutral expression, uniform illumination frontal face photo, photos people face recognition, triangular mesh decomposition combined with standard face model to generate personalized 3D face mesh, the final texture feature extraction to generate photo realistic 3D face model. In the course of the study, the face recognition algorithm, the feature extraction algorithm and model generation algorithm is the main steps of the modeling. Based on the the Sobel-Otsu human face profile recognition algorithm is a new face recognition algorithm. The algorithm uses the eight direction Sobel operator photo human face is extracted preliminary outline, and then use the the Otsu algorithm determined threshold value binarization preliminary outline FIG face contour binarization FIG. First, the proportion of the human face experience traditional division method to determine the area where each face Organ, based on CEAIP face feature extraction algorithm integral projection map and in the region, using the integral projection method, and finally through the the integral projection area Figure crest The judgment to determine the exact coordinates of each face organ, whereby the photo human facial feature information. Generation algorithm based on grid model realistic model is the research focus in recent years. The idea of ??this algorithm is using Delaunay algorithm the photos mesh, and then substituted into the modified standard face model to get personalized face 3D mesh model. And use of a surface smoothing algorithm based on three points of the grid, the smoothness of the mesh model. Finally, a realistic 3D face model reconstruction by extraction of the material in the photo. Finally, using visual studio 2005, CSGL development kit developed 3D automatic reconstruction system based on the algorithm realistic human face. Proved through the system based on a single photograph 3D realistic face model is entirely feasible, and by selecting the appropriate algorithm can achieve good results.

Related Dissertations

  1. Research and Development of Triple Channels Infrared Flame Detector,TH89
  2. Communication signal modulation recognition algorithm,TN911.7
  3. Chinese Named Entity Recognition Research Based on Discourse,TP391.43
  4. The soccer robot recognition algorithms and decision-making simulation,TP242
  5. The Study of Fingerprint Identification System Based on FPGA Hardware Realization,TP391.41
  6. Rapid Detection and Identification of SMD Components,TN407
  7. Car license plate automatic identification technology research,TP391.4
  8. Research on Speech Recognition Key Methods,TN912.3
  9. The Application in Eliminating Drift Error of Inertial Navigation System Based on Pattern Recognition Algorithm with Improved Back Propagation,TN966
  10. Research on Engineering Drawing Vectorization,TP391.41
  11. Research on Image Fire Alarm System,TP277
  12. Design of Flame Detection System Based on Multi-infrared Bands Sensor,TP274.4
  13. The fuze safety logic control circuit optimization design,TJ430.2
  14. Modern Chinese Automatic Words and several algorithms achieve,TP391.1
  15. The Research and Implementation of High Performance Embedded Real-time Image Processing and Recognition System,TP391.41
  16. Moving target feature extraction and vehicle color recognition algorithm,TP391.41
  17. The Research and Implementation of License Plate Localization and Character Segmentation Based on Embedded System,TP391.41
  18. Research and Implementation of Embedded Speech Recognition,TN912.34
  19. The Design and Application of Wireless Sensor Network Nodes,TN929.5
  20. Research and Implementation of Fingerprint Recognition Algorithm Based on ARM9 Embedded System,TP391.41

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