Deep Learning Based Face Detection Algorithm for Mobile Applications

摘要

This article proposes a face detection algorithm based on deep learning for mobile applications. Face detection is a pre-processing step for many high-end computer vision tasks. Therefore, outcomes of this task directly influence the accuracy of the other tasks, such as recognition, tracking, relighting, and other tasks. Face detection is a well studied research extent; however, it still endures from challenges such as face orientation, occlusion, lighting conditions, and alteration in facial landmarks. The projected technique aims at targeting such crucial challenges with light weight deep learning architecture. The proposed algorithm is tested on the publicly available perplexing datasets and has shown promising results. The effectiveness of the proposed algorithm is proved quantitatively and qualitatively with state of art techniques.