Similar Face Search Based On Face Cognition

The exploration of face recognition model has a long history, and has been successfully used in cosmetic surgery and other fields. Meanwhile, the exploration of facial similarity measure methods has been carried out in the field of computer vision and pattern recognition. Some applications, such as face recognition, face retrieval, and similar face search applications, have been widely used in many fields. However, the existing methods to measure face similarity don't incorporate face cognition patterns used by human beings. Therefore, the results of current methods are usually less than optimal from the perspective of human cognitive habits.Based on face cognition patterns, we define common types of each facial organ (eye, nose, mouth, facial shape) as its standard types, for example eyes' standard types containing phoenix eyes, droop eyes and so on. Then we construct standard type space for each facial organ, and map facial organs of a face image to different points in their responding standard type spaces. So we can calculate shape similarity between two facial organs of the same kind. Thereafter, we employ a circular local binary pattern (LBP) operator to calculate the local texture similarity of corresponding facial organs. The combination of shape and texture similarities is used as the criterion for similar face search. We use our method and the method provided by Face++ which presents the highest level of similar face research to test 80 frontal neutral and head angle face images respectively. The accuracy rates of our TOPI and TOP2 research results are obviously higher than Face++. Experimental results show that the search results of our method are more satisfactory from the perspective of human cognitive habits. Our proposed image search approach based on cognitive models can also be applied in the business sector, e.g. image based similarity search of online goods.