Abstract:Inthis paper, wedescribe afusion-based facerecog-Challenge (3)andgive details ofourfacerecognition method nition methodthatisabletocompensate forfacial expressions whichutilizes multiple facial regions. Insection III, we evenwhentraining samples contain onlyneutral expression. present theempirical results ofvarious methodsincluding Thesimilarity metric between twofacial images arecalculatedpeeF rical re sultfreiou methods, incudin bycombining thesimilarity scoresofthecorresponding facialtheFRGCbaselethesgle-region methods, andthemulti- regions, e.g. thesimilarity between twomouthes, thesimilarity region methods. Section IVgives adiscussion ofissues raised between twonoses, etc.. Incontrast withotherapproaches bytheapproach presented inthis paper. Finally, we make whereequal weights areassigned oneachregion, anovel fusionconcluding remarks insection V. methodbasedonlinear discriminant analysis (LDA)isdeveloped tomaximize theverification performance. We alsoconduct a II.THEDATASETANDTHEPROPOSEDMETHOD comparative study onvarious facerecognition schemes, including A TheFRGC v2.0dataset theFRGC baseline algorithm, thefusion ofmultiple regions by sumrule, andthefusion ofmultiple regions byLDA.ExperimentsWeperform facerecognition experiments ontheFRGCv2.0 ontheFRGC (Face Recognition GrandChallenge) V2.0dataset, dataset (3). IntheFRGCdataset, 3Dimages witharesolution containing 4007faceimages recorded from266subjects, show 640x 480consist ofbothshape andtextures channels. The thattheproposed methodsignificantly improves theverification performance inthepresence offacial expressions. twodisjoint datapartitions, training andvalidation partitions, contains 943and40073Dimages respectively. Amongthe

A Fusion-Based Method for 3D Facial Gender Classification

Fusion of Multiple Facial Regions for Expression-Invariant Gender Classification

A Novel Fusion-Based Method for Expression-Invariant Gender Classification

A new facial feature based on the fusion of texture and shape characteristics

Gender Classification of Facial Images Based on Multiple Facial Regions

High-Resolution Face Fusion for Gender Conversion

Gender categorization based on 3D faces

An Efficient 3D Face Recognition Approach Based on the Fusion of Novel Local Low-Level Features

Gender Recognition Based on Face Features

Sexual Dimorphism Analysis and Gender Classification in 3D Human Face.

Enhancing gender classification by combining 3D and 2D face modalities

Automatic facial expression recognition on a single 3D face by exploring shape deformation.

Fusion of Multiple Facial Regions for Expression-Invariant Face Recognition

Gender Classification Using 3D Statistical Models

Fusing Normal Vector and Curvature Features on the Mesh for 3D Facial Expression Recognition.

Gender Identification Based on the Fusion of Adaboost and SVM

Facial Expression Recognition Based On Features Fusion

Gender recognition from visible and thermal infrared facial images

Fusion Ofmultiple Facial Regions For

A Novel Race Classification Method Based on Periocular Features Fusion

Gender Classification by Combining Facial and Hair Information