Photo-realistic Face Analysis and Synthesis in Image and Video to Generate Personal Characteristics

2016 Fiscal Year Final Research Report

• Project/Area Number: 26280062
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Perceptual information processing
• Research Institution: Waseda University
• Principal Investigator: Morishima Shigeo 早稲田大学, 理工学術院, 教授 (10200411)
• Project Period (FY): 2014-04-01 – 2017-03-31
• Keywords: 3次元顔形状復元 / リップシンク合成 / 経年変化顔動画像合成 / 似顔絵からの実写復元 / キャラクタ合成 / 肥痩顔画像合成 / 犯罪捜査支援システム / パッチベース画像合成

Outline of Final Research Achievements

This research project has succeeded to produce 6 systems concerning to face image analysis and synthesis considering personal characteristics as follows. 1) Face montage system effective to criminal investigation by considering an aging and slim/fat effect handling. 2) High quality pixel by pixel 3D face geometry reconstruction by single shot image with only small database. 3) Video lip synchronization system by video re-shuffling method applicable to translated voice without any 3D model for entertainment application. 4) High quality instant casting system for dance character reflecting personal characteristics in face and body. 5) Aging face video generation system reflecting personal characteristics in expression transition. 6) Automatic photo-realistic image generation from facial caricature keeping personal feature in original image.

Free Research Field

画像情報処理