Shape-appearance-correlated active appearance model. (August 2016)

• Record Type: Journal Article
• Authors: Zhou, Huiling; Lam, Kin-Man; He, Xiangjian
• Is Part Of: Pattern recognition. Volume 56(2016:Aug.)
• Page Start: 88
• Page End: 99
• Publication Date: 2016-08
• Subjects: Facial-feature localization -- Generic active appearance model -- Canonical correlation analysis -- Orthogonal CCA; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2016.03.002
• Languages: English
• ISSNs: 0031-3203

Abstract

Abstract: Among the challenges faced by current active shape or appearance models, facial-feature localization in the wild, with occlusion in a novel face image, i.e. in a generic environment, is regarded as one of the most difficult computer-vision tasks. In this paper, we propose an Active Appearance Model (AAM) to tackle the problem of generic environment. Firstly, a fast face-model initialization scheme is proposed, based on the idea that the local appearance of feature points can be accurately approximated with locality constraints. Nearest neighbors, which have similar poses and textures to a test face, are retrieved from a training set for constructing the initial face model. To further improve the fitting of the initial model to the test face, an orthogonal CCA (oCCA) is employed to increase the correlation between shape features and appearance features represented by Principal Component Analysis (PCA). With these two contributions, we propose a novel AAM, namely the shape-appearance-correlated AAM (SAC-AAM), and the optimization is solved by using the recently proposed fast simultaneous inverse compositional (Fast-SIC) algorithm. Experiment results demonstrate a 5–10% improvement on controlled and semi-controlled datasets, and with around 10% improvement on wild face datasets in terms of fitting accuracy compared to other state-of-the-art AAM models. Highlights: A fast face-model initialization scheme is proposed for AAMs. Orthogonal CCA is employed to increase the correlation between shape and appearance. A shape-appearance-correlated AAM is proposed to tackle the generic problem of localizing feature points on unseen faces.