Research Paper on Expressive Facial Animation Synthesis by Learning Speech Co-articulation and Expression Spaces

Abstract

In the current project, the techniques and different algorithms of the researchers are mentioned in the paper. The introduction is mentioned along with the common method to identify the facial expression which is also discussed in the paper. The literature review contains the historical researches by the authors regarding the 3D visualization and facial expression resignation. It was purposed to find the three different facial expressions of a person by dividing the face of a person into six sub-regions. The range of the markers was selected to be between 90 and 102. According to the markers or sub-region on the face, the expressions were identified which were red for the anger, green for the happiness and blue for the sadness. According to the data obtained and the results, it is identified and concluded that the 3D representation of the facial expression is more accurate than the single line representation.