Human Face Reconstruction using Divine Proportions and Gestalt for Occluded Video Face Recovery in Forensic Analysis using Deep Learning

Forensic video analysis has been used in diverse kind of high-profile cases, global discrepancies, and conflict zones. It is a three-phase process of scientific examination, comparison, and evaluation of video in legal matters. Human face reconstruction using deep learning for occluded video face recovery to aid in forensic analysis is the main objective of this paper. Forensic facial reconstruction is a combination of both scientific methods and artistic skill. In this paper, we introduce a method to reconstruct human faces occluded due to short noise innight-time video clips. A skull database is created with unique skull models with varying shapes, forms and proportions. Human body mathematical model biometric using golden ratio algorithm is proposed and used to find the occluded face proportions. Closure principle of gestalt theory of visual perception is used to fill in the missing parts of a face design and to create a whole face image using gan. The proposed model is found to have 50% lesser reduced Median error rate and 20% reduced Stdev than PrNet and 10% lower Mean error rate than 3Dddfav2.