Optimizable face detection and tracking model with occlusion resolution for high quality videos

Recent state-of-the-art Face Detection algorithms in the field of Computer Vision focus greatly on real-time processing and results. The applications using these algorithms deal with low quality video feeds having less Pixels Per Inch (ppi) and/or low frame rate. The algorithms perform well with such video feeds, but their performance deteriorates towards high quality, high data-per-frame videos. Such video files mostly exist in offline mode, that could be used for post processing by the Computer Vision applications. This paper focuses on developing such an algorithm that gives faster results on high quality videos, at par with the algorithms working on live low quality video feeds. The proposed algorithm uses Convolutional-MTCNN as base algorithm, and speeds it up for high definition videos. This paper also presents a novel solution to the problem of occlusion and detecting partial or fully hidden faces in the videos. This is achieved by using probabilistic approaches, given that the face has been identified in first few frames, to give the algorithm an estimate of where the face should be in the occluded region.

Automated summary

This paper presents a face detection algorithm that achieves faster results in high quality videos. The algorithm uses Convolutional-MTCNN as the base algorithm and solves the problem of occlusion by using probabilistic approaches.