3D　face　database　is　an　important　data　platform　for　model　training,　algorithm　design.　Especially　for　these　researches　based　on　statistical　learning　method,　the　algorithm　performance　has　a　great　dependency　on　training　sample　set.　Subject　to　the　constraint　of　data　acquisition　equipment　the　data　structure　of　current　3D　face　samples　are　different.　It　is　incontinent　to　use　these　samples　without　regulation.　To　solve　this　problem,　this　paper　presents　an　regulating　method　for　existing　3D　databases　based　on　surface　deformation.　The　experiment　results　show　that　the　proposed　method　has　good　performance　on　face　regulation.