Single image deblurring for a real-time face recognition system

Abstract

Abstract-Blur due to motion and atmospheric turbulence is a variable that impacts the accuracy of computer vision-based face recognition techniques. However, in images captured in the wild, such variables can hardly be avoided, requiring methods to account for these degradations in order to achieve accurate results in real time. One such method is to estimate the blur and then use deconvolution to negate or, at the very least, mitigate the effects of blur. In this paper, we describe a method for estimating motion blur and a method for estimating atmospheric blur. Unlike previous blur estimation methods, both methods are fully automated, allowing integration into a real-time facial recognition pipeline. We show experimentally, on datasets processed to include synthetic and real motion and atmospheric blur, that these techniques improve recognition more than prior work. At multiple levels of blur, our results demonstrate significant improvement over related works and our baseline on data derived from both the FERET (fairly constrained data) and Labeled Faces in the Wild (fairly unconstrained data) sets.

Top-down facilitation of multistage decisions for face recognition

“…It is interesting to note that the classes whose models do not provide a score in the top 10 during the first classification phase seem only to confuse the second classification phase, which suggests that reducing the digital and optical resolution of a degraded image may provide a more accurate classification than attempting to compare a degraded probe against clean gallery images. While these results represent marked improvement over the baseline blurred recognition results, other attempts to solve the blur problem such as [10] demonstrates an equivalent or better performance increase on the same data via the simple procedure of deblurring and recognizing using a single unmodified SVM.…”

“…However, this discrepancy can be made up by deblurring the original input probe using a Wiener deconvolution with a point spread function generated from the blur parameters as described in [10]. Since this study is not on deblurring methods, we simply use the ground truth blur parameters obtained when generating the datasets.…”

Face and Body-Based Human Recognition by GAN-Based Blur Restoration

“…Face recognition was performed using PCA, linear discriminant analysis (LDA), kernel principal component analysis (KPCA), and kernel Fisher analysis (KFA). Heflin et al [ 15 ] used the FERET database wherein the face area was detected in the blurred image, motion blur and atmospheric blur were measured using a blur point spread function (PSF), and, finally, face deblurring was performed using a deconvolution filter, such as Wiener filter, to evaluate the recognition performance. Yasarla et al [ 16 ] proposed uncertainty guided multi-stream semantic network (UMSN) and performed facial image deblurring.…”

A Look at Eye Detection for Unconstrained Environments

“…The problem with this approach is that under truly difficult conditions, as opposed to the very controlled settings of [27] (full frontal imagery, with a constant inter-ocular distance), it is likely that a collection of detected faces in a direct temporal sequence will not be possible, thus reducing the potential of such algorithms. Real-time techniques to recover facial images degraded by motion and atmospheric blur were explored in [9]. The experiments of [9] with standard data sets and live data captured at 100m showed how even moderate amounts of motion and atmospheric blur can effectively cripple a facial recognition system.…”

“…Real-time techniques to recover facial images degraded by motion and atmospheric blur were explored in [9]. The experiments of [9] with standard data sets and live data captured at 100m showed how even moderate amounts of motion and atmospheric blur can effectively cripple a facial recognition system. The work of [4] and [9] is more along the lines of what is explored in this paper, including a thorough discussion of the underlying issues that impact algorithm design, as well as an explanation of how to perform realistic controlled experiments under difficult conditions, and algorithmic issues such as predicting when a recognition algorithm is failing in order to enhance recognition performance.…”