Fast and exact Newton and bidirectional fitting of Active Appearance Models

Kossaifi, Jean; Tzimiropoulos, Georgios; Pantic, Maja

Abstract

Active Appearance Models (AAMs) are generative models of shape and appearance that have proven very attractive for their ability to handle wide changes in illumination, pose and occlusion when trained in the wild, while not requiring large training dataset like regression-based or deep learning methods. The problem of fitting an AAM is usually formulated as a non-linear least squares one and the main way of solving it is a standard Gauss-Newton algorithm. In this paper we extend Active Appearance Models in two ways: we first extend the Gauss-Newton framework by formulating a bidirectional fitting method that deforms both the image and the template to fit a new instance. We then formulate a second order method by deriving an efficient Newton method for AAMs fitting. We derive both methods in a unified framework for two types of Active Appearance Models, holistic and part-based, and additionally show how to exploit the structure in the problem to derive fast yet exact solutions. We perform a thorough evaluation of all algorithms on three challenging and recently annotated inthe- wild datasets, and investigate fitting accuracy, convergence properties and the influence of noise in the initialisation. We compare our proposed methods to other algorithms and show that they yield state-of-the-art results, out-performing other methods while having superior convergence properties.

Citation

Kossaifi, J., Tzimiropoulos, G., & Pantic, M. (2017). Fast and exact Newton and bidirectional fitting of Active Appearance Models. IEEE Transactions on Image Processing, 26(2), 1040 - 1053. https://doi.org/10.1109/TIP.2016.2642828

• Journal Article Type: Article
• Acceptance Date: Dec 5, 2016
• Online Publication Date: Dec 21, 2016
• Publication Date: Feb 28, 2017
• Deposit Date: Jan 16, 2017
• Publicly Available Date: Jan 16, 2017
• Journal: IEEE Transactions on Image Processing
• Print ISSN: 1057-7149
• Electronic ISSN: 1941-0042
• Publisher: Institute of Electrical and Electronics Engineers
• Peer Reviewed: Peer Reviewed
• Volume: 26
• Issue: 2
• Pages: 1040 - 1053
• DOI: https://doi.org/10.1109/TIP.2016.2642828
• Keywords: Active Appearance Models, Newton method, bidirectional image alignment, inverse compositional, forward additive
• Public URL: https://nottingham-repository.worktribe.com/output/832636
• Publisher URL: http://ieeexplore.ieee.org/document/7792677/
• Additional Information: (c) 2016 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other users, including reprinting/ republishing this material for advertising or promotional purposes, creating new collective works for resale or redistribution to servers or lists, or reuse of any copyrighted components of this work in other works