FAP BAP blender - isir/greta GitHub Wiki
FAP/BAP blender
In order to use multiple behavior generators in parallel, we created FAP/BAP blenders.
- Since this modules assume the input signals are two, you need to cascade blenders to blender signals more than three.
- When you connect modules, you need to specify the corresponding input node names (check the following screenshot as an example).
- You must assign different node name for each input signal otherwise it does not work properly.
- Please note that the names can be different for each blender.
Greta has the following frame-level representations:
- FAPFrame: frame-level representation of facial expressions
- BAPFrame: frame-level representation of body expressions
For FAPFrame, we implemented the following blenders:
- LipBlender
- FaceBlender
For BAPFrame, we implemented the following blenders:
- BodyBlender
For more details, please check the following implementation files:
- LipBlender: greta\core\MPEG4\src\greta\core\animation\mpeg4\fap\filters\LipBlender.java
- FaceBlender: greta\core\MPEG4\src\greta\core\animation\mpeg4\fap\filters\FaceBlender.java
- BodyBlender: greta\core\MPEG4\src\greta\core\animation\mpeg4\bap\filters\BodyBlender.java