This folder contains the data which has been used for facial animation by Kang Liu in the year 2009. This data was used in a project to generate a talking head, which was funded by the DFG.
The data contains of 145 sentences which are described in the file "corpus.txt" of the main folder.

COPYRIGHT: 
Leibniz University Hannover (LUH), Institut fuer Informationsverarbeitung (TNT), Prof. Dr.-Ing. J. Ostermann
When using the provided data, please refer to one of our papers:
- K. Liu, J. Ostermann, “Evaluation of an Image-based Talking Head with Realistic Facial Expression and Head Motion”, Journal on Multimodal User Interfaces, Special issue: Emotion-based Interaction, Springer Verlag, October 2011
-  K. Liu and J. Ostermann, “Realistic facial expression synthesis for an image-based talking head,” in IEEE Conference on Multimedia and Expo, ICME2011, Jul. 2011, p. 6.

_________________________________________
The provided data consists of three components:
(1) Video: yuv
(2) Audio: wav
(3) Phonetic information: phoneme
These different data parts are separated in three different folders, which will be shortly described in the following.
Additionally a textfile "corpus.txt" is provided which contains the actual text corpus which has been read by the actor.

______________________________
(1) Video Data in folder "yuv"
______________________________
  contains *.yuv files with 145 short sequences. 
  Each filename contains two numbers: first three digits give information about the sequence number ranging from 1 to 145, second integer number ranging from 1 to 3, indicates the expression of the mouth, where:
  1=neutral, 2=smile while speaking, 3=smile at the end
  They can be viewed by the following command line:
  mplayer -demuxer rawvideo -rawvideo w=576:h=720:fps=50 david_001_1.yuv
  recommaneded view with scaled version:
  mplayer -demuxer rawvideo -rawvideo w=576:h=720:fps=50 -xy 2 david_001_1.yuv
______________________________
(2) Audio Data in folder "wav"
______________________________
  contains the audio files which correspond to the video data of the folder yuv

______________________________
(3) Phonetic information in folder "phoneme"
______________________________
  contains *.txt files, where the names correspond to the files in the folder "mouth"
  each file contains the phonetic information of the corresponding mouth sequence, where the first columns is the duration in miliseconds and the second column contains the name of the phoneme
______________________________