In order to meet the speed and accuracy requirements of face key point detection (face alignment) in ap-plication scenarios, firstly, cascaded prediction is carried out on the basis of SSD (single shot multibox detector),which combines more uniformly distributed feature layers to form MR-SSD (more robust SSD), a deep learning de-tector with more robust response to multi-scale faces. Secondly, based on the cascade shape regression method oflocal binary feature (LBF), a multi-angle initialization algorithm based on the difference between the facial pixels isproposed. Five groups of feature points in the 90 degree inclination range of positive and negative face are initializedto achieve excellent fitting effect for inclined face under multi angles. The mean square deviation of each group offeature points after regression is calculated and the maximum corresponding shape is used as the final regressionshape. The optimal architecture proposed in this paper can obtain robust face bounding box and face alignmentschemes against multi-angle tilt in real time.