Data
Facial Motion Capture Data
The mocap data are used for segmentation, registration, and compression, i.e. all related techniques for processing mesh sequences.
- Facial expressions
The data contains the following expressions: Eyebrow raising, Anger, Disgust, Fear, Happiness, Surprise and Sadness.
- Lip movements
The subject says the following sentences/words :
- "The rain in spain mainly stays in the plaine.”
- “How kind of you to let me come.”
- Do->Re->Mi->Fa->So->La->Ti
- One->Two->Three->... Ten.
- Asymmetric expressions
The data contains the following facial actions: Left-eye wink, both eyes blinking, right-eye wink, left-side smile, right-side smile.
Deforming Meshes
A deforming mesh is a sequence of meshes with fixed topology, i.e. the mesh has fixed number vertices and connectivity. The following deforming meshes are grouped based on motions. All the data are available in '.max' format.
- Facial expressions
We transfer the previously captured facial expressions to the scanned faces of the corresponding person.
- Walk
These data are obtained by rigging TOSCA high-resolution meshes to an identical walking skeleton provided by 3dsmax Studio.
- Gallop
We obtain the Galloping 'Camel' and 'Horse_1' from Sumner et al.’s work on Deformation Transfer http://people.csail.mit.edu/sumner/research/deftransfer/data.html#download. Courtesy of Arash Habibi, Horse_2 is also a Galloping animation, but with different frame rate, different starting pose and different duration.
Contact
The data can be used under the following terms and conditions:
- The data can be used for research purposes only.
- The data should not be used for commercial purposes.
- The data can be granted that the source of the data is mentioned.
In order to get a permission to access to these data, please email a request to Hyewon SEO with seo AT unistra.fr. Intranet:Mocap