Face　detection　is　a　classical　problem　in　the　field　of　computer　vision.　It　is　widely　used　in　recent　years,　face　detection　and　face　tracking　has　not　only　limited　to　the　scope　of　application　of　face　recognition:　in　video　retrieval,　video　surveillance,　facial　expression　analysis,　gender,　race,　age　discrimination,　digital　entertainment,　and　so　on.　This　paper　proposed　algorithm　based　on　AdaBoost　algorithm　AAM　model　of　face　feature　points　to　identify　the　improvement　in　a　certain　range　to　solve　the　present　stage　AAM　algorithm　does　not　consider　the　grayscale　in　the　exact　face　of　initial　position　and　face　face　detection　problem.　©　Springer　Nature　Singapore　Pte　Ltd.　2017.