Unconstrained Face Detection and Open-Set Face Recognition Challenge

Abstract

Face detection and recognition benchmarks have shifted toward more difficult environments. The challenge presented in this paper addresses the next step in the direction of automatic detection and identification of people from outdoor surveillance cameras. While face detection has shown remarkable success in images collected from the web, surveillance cameras include more diverse occlusions, poses, weather conditions and image blur. Although face verification or closed-set face identification have surpassed human capabilities on some datasets, open-set identification is much more complex as it needs to reject both unknown identities and false accepts from the face detector. We show that unconstrained face detection can approach high detection rates albeit with moderate false accept rates.By contrast, open-set face recognition is currently weak and requires much more attention.

Researchers Gone Wild

“…TheUCCSdatasetincludesover1,700uniqueidentitiesofstudentsandfacultywalkingtoandfromclass.Thephotosweretaken during the spring semesters of school year 2012-2013 on the West LawnoftheUniversityofColorado,ColoradoSpringscampus,using aCanon7D18-megapixeldigitalcamerafittedwithaSigma800mm F5.6EXAPODGHSMtelephotolens,pointedoutanofficewindow acrosstheuniversity'sWestLawn. "Thecamera[was]programmed to start capturing images at specific time intervals between classes to maximize the number of faces being captured" (Günther et al 2017). Their setup made it impossible for students to know they were being photographed, providing the researchers with realistic, unconstrained, surveillance images to help build face recognition systemsforrealworldapplicationsbydefense,intelligenceandcommercialpartners.Infact,thedatasetwasfundedbytheIntelligence AdvancedResearchProjectsActivity(IARPA),theOfficeofDirector ofNationalIntelligence(ODNI),OfficeofNavalResearchandtheDe-partmentofDefenseMultidisciplinaryUniversityResearchInitiative (ONRMURI),andtheSpecialOperationsCommandandSmallBusi-nessInnovationResearch(SOCOMSBIR).AUniversityofColorado, ColoradoSpringswebsitealsoexplicitlystatesthattheirinvolvement intheIARPAJanusfacerecognitionprojecthasbeendevelopedto servetheneedsofnationalintelligence,establishingthatthedataset of student images was created in the interest of United States defenseandintelligenceagencies.…”

Neural Network Explainable AI Based on Paraconsistent Analysis: An Extension

“…The open world recognition problem is intrinsically problematic due to the infinity property of open sets. Some attempts at handling this problem range between early approaches, such as the use of a threshold [18] and a "garbage" (or "background") class [19], to more recent ones, such as the open-max, which uses the extreme value theory within a neural network for calibrating the compact abating probability [20], and an approach that identifies unknown classes through a proposed loss function [21]. To the best of the author's knowledge, this is the first effort of using annotated paraconsistent analysis as an XAI alternative for addressing the open set issue in neural networks.…”

Towards a Reliable Face Recognition System

“…Apart from generalizing better, deep learning methods enhance performances through preprocesses such as augmentation, random cropping, hard mining of samples, negative detection and others. Günther et al [10] observed that on open-set detection challenge using UCCS dataset, both TinyFaces, Cascade CNN, YOLO, LBF and LgfNet performed well on face detection. The models were able to detect at least 33000 of the 36153 labeled test faces.…”