LEARNING GEOMETRIC 3D FACIAL MOTION MODEL - 3D Face Processing: Modeling, Analysis and Synthesis - page 24

The learned parts-based MUs give more flexibility in local facial deformation
analysis and synthesis. Figure 3.5 and 3.6 show some local deformation in lower
lips and right cheek, each of which is induced by one of the learned parts-based
MUs. These locally deformed shapes are difficult to approximate using holistic
MUs in Section 3.

For each deformation shown in Figure 3.5 and 3.6, more

Figure 3.5. Three lower lips shapes deformed by three of the lower lips parts-based MUs
respectively. The top row is the frontal view and the bottom row is the side view.

Figure 3.6. (a): The neutral face side view. (b): The face deformed by one right cheek parts-

based MU.

than 100 holistic MUs are need to achieve a 90% reconstruction accuracy. That
means, although some local deformation is induced by only one parts-based
MU, more than 100 holistic MUs may be needed in order to achieve good
analysis and synthesis quality. Therefore, we can have more flexibility in using
parts-based MUs. For example, if we are only interested in motion in forehead,