A NEW WAY TO DEFINE FACES DEPENDS ON MULTI-METRIC GRAPHS

Abstract

Facial detection plays an important role in many applications, such as video surveillance, sex classification, facial recognition. In this paper, we present a new way of determining the face based on a multi-faceted graph. The proposed method uses a multi-level graph to represent the face, so as to improve computational efficiency, making the procedure suitable for large data multimedia databases. While most of the current methods focus on indexing high-dimensional visual features and also focus on scalability limits, using this system to the peer system, image-based retrieval of content in the pouch system to a partner is feasible. Word templates should update the notebook periodically during such an atmosphere, rather than installing storage. Within this paper, we introduce the universal coded system as a single dynamic generation approach, which considers the balance between both discrimination and workload. In addition, dynamic peer-to-peer networks often evolve, and become less stable to retrieve a fixed codebook. In order to improve recovery performance and lower network costs, the printing technology index has been developed. Unlike central environments, the main challenge is that you will be able to get an efficient global codec, such as images distributed over peer-to-peer networks.