Automatic 3D Face Model Adaptation with Two Complexity Modes for Visual Communication

2011 ◽  
pp. 295-316
Author(s):  
Markus Kampmann ◽  
Liang Zhang
Keyword(s):  
Visual Communication ◽  
Low Complexity ◽  
Facial Features ◽  
Model Adaptation ◽  
High Complexity ◽  
Face Model ◽  
3D Face ◽  
Automatic Adaptation ◽  
Complete Framework ◽  
Mode A

This chapter introduces a complete framework for automatic adaptation of a 3D face model to a human face for visual communication applications like video conferencing or video telephony. First, facial features in a facial image are estimated. Then, the 3D face model is adapted using the estimated facial features. This framework is scalable with respect to complexity. Two complexity modes, a low complexity and a high complexity mode, are introduced. For the low complexity mode, only eye and mouth features are estimated and the low complexity face model Candide is adapted. For the high complexity mode, a more detailed face model is adapted, using eye and mouth features, eyebrow and nose features, and chin and cheek contours. Experimental results with natural videophone sequences show that with this framework automatic 3D face model adaptation with high accuracy is possible.

scholarly journals Distinguishing Planting Structures of Different Complexity from UAV Multispectral Images

Sensors ◽  
2021 ◽  
Vol 21 (6) ◽  
pp. 1994
Author(s):  
Qian Ma ◽  
Wenting Han ◽  
Shenjin Huang ◽  
Shide Dong ◽  
Guang Li ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Confusion Matrix ◽  
Object Oriented ◽  
Low Complexity ◽  
Classification Model ◽  
Recursive Feature Elimination ◽  
Support Vector ◽  
Multispectral Images ◽  
High Complexity ◽  
Structure Complexity

This study explores the classification potential of a multispectral classification model for farmland with planting structures of different complexity. Unmanned aerial vehicle (UAV) remote sensing technology is used to obtain multispectral images of three study areas with low-, medium-, and high-complexity planting structures, containing three, five, and eight types of crops, respectively. The feature subsets of three study areas are selected by recursive feature elimination (RFE). Object-oriented random forest (OB-RF) and object-oriented support vector machine (OB-SVM) classification models are established for the three study areas. After training the models with the feature subsets, the classification results are evaluated using a confusion matrix. The OB-RF and OB-SVM models’ classification accuracies are 97.09% and 99.13%, respectively, for the low-complexity planting structure. The equivalent values are 92.61% and 99.08% for the medium-complexity planting structure and 88.99% and 97.21% for the high-complexity planting structure. For farmland with fragmentary plots and a high-complexity planting structure, as the planting structure complexity changed from low to high, both models’ overall accuracy levels decreased. The overall accuracy of the OB-RF model decreased by 8.1%, and that of the OB-SVM model only decreased by 1.92%. OB-SVM achieves an overall classification accuracy of 97.21%, and a single-crop extraction accuracy of at least 85.65%. Therefore, UAV multispectral remote sensing can be used for classification applications in highly complex planting structures.

scholarly journals Detail Reconstruction of 3D Face Model from Single Image

2021 ◽  
Vol 1948 (1) ◽  
pp. 012053
Author(s):  
Hongxin Xu ◽  
Ruoming Lan ◽  
Tianping Li
Keyword(s):  
Single Image ◽  
Face Model ◽  
3D Face

Matching Forms as a Function of Complexity, Type of Match, and Retention Interval

1976 ◽  
Vol 43 (3_suppl) ◽  
pp. 1299-1302
Author(s):  
Virginia Brabender ◽  
Christopher Clay
Keyword(s):  
Retention Interval ◽  
Visual Processing ◽  
Present Experiment ◽  
Low Complexity ◽  
Stimulus Complexity ◽  
High Complexity

The present experiment tested the hypothesis that nominal processing increases as stimulus complexity increases. Subjects indicated whether two 4- or 12-sided forms, separated by an interval of .5 or 4.0 sec., were the same or different. “Same” responses corresponded to matches for physical or nominal identity. Longer RTs for high complexity than low complexity forms suggest that complexity affects the efficiency of visual processing rather than the occurrence of nominal processing. An interaction between type of match and interval, due to the longer RTs for matches of nominally identical forms at only the .5-sec. interval, indicates that at this interval, matches for physical and nominal identity are made with visual and nominal representations respectively.

Sign in / Sign up

Export Citation Format

Share Document