View source for IC 8 Python API:Facial Animation

{{TOC}}

{{Parent|IC_8_Python_API|iC8 Python API}}

iClone 8 has redesigned the facial animation architecture to allow for different expression sets and customized expression morphs. Currently, three expression sets are supported: Traditional, CC4 Standard, and CC4 Extended. You can view their differences by operating the '''Face Key''' dialog window.

[[File:ic8_python_api_facial_animation_01.png|frame|Traditional iC7 standard morphs x143]]

[[File:ic8_python_api_facial_animation_02.png|frame|CC4 standard morphs x75]]

[[File:ic8_python_api_facial_animation_03.png|frame|CC4 extended morphs x160+ (Expressions can be customized)]]

The name of the expression set can be obtained with the '''RIFaceComponent.GetExpressionSetUid()''' function.

<syntaxhighlight lang="python" line='line'>
face_component = avatar.GetFaceComponent()
print(face_component.GetExpressionSetUid())
# IC7 Standard, CC4 Standard, CC4 Extended
</syntaxhighlight>

You can also use '''RIFaceComponent.GetExpressionGroups()''' to get the category names of expressions (e.g. "Brow", "Eye", "Nose"). To get all the expression names, use '''RIFaceComponent.GetExpressionNames()''', or enter the group name to get all the expression names that belong to it. Afterward, the expression name can be used to set keys.

<syntaxhighlight lang="python" line='line'>
face_component = avatar.GetFaceComponent()
exp_groups = face_component.GetExpressionGroups()
print("expression groups="+str(exp_groups ))
# Brow, Eye, Nose…

# all expressions
exp_names = face_component.GetExpressionNames("")
print("expression names="str(exp_names))
# Brow_Raise_Inner_Left, Brow_Raise_Inner_Right, Brow_Raise_Outer_Left… 

# expressions in Eye group
exp_names = face_component.GetExpressionNames("Eye")
print("Eye expression names="str(exp_names))
# Eyes_Blink, Eye_Blink_L, Eye_Blink_R… 
</syntaxhighlight>

When you are generating facial animation, use '''RIFaceComponent.AddClip()''' to add an expression clip at a specified time. Then use '''RIFaceComponent.AddExpressionKeys''' to enter the Key in the clip range. In order to improve the performance for setting keys, '''AddExpressionKeys''' needs to be wrapped between '''RIFaceComponent.BeginKeyEditing()''' and '''RIFaceComponent.EndKeyEditing()'''. The code is as follows:

<syntaxhighlight lang="python" line='line'>
face_component = avatar.GetFaceComponent()

# get Expression names
exp_names = face_component.GetExpressionNames("")
print ("expression names="+str(exp_names))

# create an expression clip
srt_time = RLPy.IndexedFrameTime(30, RLPy.RGlobal.GetFps())
clip_length = RLPy.IndexedFrameTime(100, RLPy.RGlobal.GetFps())
face_component.AddClip(srt_time, "ExpressionClip", clip_length)

# add Expression Key
exp_val = [1.0]*len(exp_names)
Inv_time = RLPy.IndexedFrameTime(0, RLPy.RGlobal.GetFps())
key_time = RLPy.IndexedFrameTime(40, RLPy.RGlobal.GetFps())    
face_component.BeginKeyEditing()
face_component.AddExpressionKeys( key_time, exp_names, exp_val, Inv_time )
face_component.EndKeyEditing()
</syntaxhighlight>

If you plan on setting keys at a fixed time interval, you can also specify '''Inv_time''' when '''RIFaceComponent.AddExpressionKeys''' is used to quicken the process.

<syntaxhighlight lang="python" line='line'>
face_component = avatar.GetFaceComponent()

# get Expression names
exp_names = face_component.GetExpressionNames("")
print ("expression names="+str(exp_names))

# create an expression clip
srt_time = RLPy.IndexedFrameTime(30, RLPy.RGlobal.GetFps())
clip_length = RLPy.IndexedFrameTime(100, RLPy.RGlobal.GetFps())
face_component.AddClip(srt_time, "ExpressionClip", clip_length)

# add Expression Keys repeatedly
exp_val = [0.0]*len(exp_names)
exp_val += [0.5]*len(exp_names)
exp_val += [1.0]*len(exp_names)
exp_val += [0.5]*len(exp_names)
exp_val += [0.0]*len(exp_names)

Inv_time = RLPy.IndexedFrameTime(10, RLPy.RGlobal.GetFps())
key_time = RLPy.IndexedFrameTime(30, RLPy.RGlobal.GetFps())    
face_component.BeginKeyEditing()
face_component.AddExpressionKeys( key_time, exp_names, exp_val, Inv_time )
face_component.EndKeyEditing()
# add 0.0 at Frame 30
# add 0.5 at Frame 40
# add 1.0 at Frame 50
# add 0.5 at Frame 60
# add 0.0 at Frame 70
</syntaxhighlight>

In addition, if you use mocap to generate facial animation, you'll need to make corresponding changes to call the API via '''RIFaceComponent.GetExpressionNames("", True)'''. The second parameter must be set to "True" to get a list of expression names that are ordered according to the mocap. When calling '''RIFacialDevice.ProcessData()''', the expression values in this order must be substituted. An example of this in practice is as follows:

<syntaxhighlight lang="python" line='line'>
face_component = mocap_avatar.GetFaceComponent()

exp_names = face_component.GetExpressionNames("", True)
print(str(exp_names))

# start mocap
mocap_manager.Start(RLPy.EMocapState_Preview)

# after receiving facial data from the device…
exp_list = [0.5]*len(exp_names)
facial_device.ProcessData(mocap_avatar, exp_list)

# end mocap
mocap_manager.Stop()
</syntaxhighlight>