an improved 3d face modeling method based on morphable model 改進(jìn)的基于形變模型的三維人臉建模方法
a 3d morphable model based on components is present in the thesis 提出了基于部件的三維形變模型。
1 romdhani s, blanz v, vetter t . face identification by fitting a 3d morphable model using linear shape and texture error functions . in proc . 7th european conf 并且,通過(guò)分塊仿射變換得到的紋理已經(jīng)損失了形狀的信息,紋理和形狀的相關(guān)性已經(jīng)減弱,過(guò)強(qiáng)的假設(shè)使得結(jié)果的推廣性能減弱。
however, the pose and illumination variation is still very challenging for face recognition . our work concentrates on the 3d morphable model and recognizing faces from different poses 但非理想成像條件下(如光照條件不理想,人臉的姿態(tài)發(fā)生變化等)的人臉識(shí)別技術(shù)還遠(yuǎn)未成熟,有許多科學(xué)問(wèn)題需要研究。
in the process of matching a morphable model to image, the euclidean distance between the reconstructed 3d component image and the input component image is defined as the goal function . comparing with the whole face 3d morphable model, the dimension of each component is low and the scale of the goal function of each component is relatively small, so the optimized result is improved and the 3d component is more precise 在模型匹配過(guò)程中,定義相同視點(diǎn)下獲得的三維部件圖像與給定部件圖像的誤差作為目標(biāo)函數(shù),由于每個(gè)部件維數(shù)比較低,與整個(gè)面部的三維形變模型相比,目標(biāo)函數(shù)的規(guī)模相對(duì)較小,優(yōu)化效果得到改善,提取的三維部件特征更準(zhǔn)確。