Efficient Deep Face Alignment with Explicit Statistical Shape Models in Convolutional Neural Networks
Kopaczka M, Schock J, Kruse P, Merhof D (2020)
Publication Type: Conference contribution
Publication year: 2020
Publisher: Institute of Electrical and Electronics Engineers Inc.
Conference Proceedings Title: 2020 10th International Conference on Image Processing Theory, Tools and Applications, IPTA 2020
Event location: Virtual, Paris, FRA
ISBN: 9781728187501
DOI: 10.1109/IPTA50016.2020.9286671
Abstract
Face alignment is commonly solved by utilizing a shape representation which is either implicitly learned as in current deep architectures or explicitly given as in active shape models and their derivatives. In this work, we present a method for integrating explicit statistical shape priors into a deep convolutional network by introducing a novel layer which allows using statistical PCA-based models directly in the network, resulting in a highly efficient deep face alignment architecture capable of performing landmark detection with several hundreds of frames per second on consumer hardware. In addition to introducing the core algorithm, we also introduce and discuss extensions for improving fitting precision. We validate the performance of all methods quantitatively on the 300W dataset, where we achieve state-of-the-art precision while requiring only a fraction of the runtime of most competing methods.
Involved external institutions
Germany (DE)How to cite
APA:
Kopaczka, M., Schock, J., Kruse, P., & Merhof, D. (2020). Efficient Deep Face Alignment with Explicit Statistical Shape Models in Convolutional Neural Networks. In 2020 10th International Conference on Image Processing Theory, Tools and Applications, IPTA 2020. Virtual, Paris, FRA: Institute of Electrical and Electronics Engineers Inc..
MLA:
Kopaczka, Marcin, et al. "Efficient Deep Face Alignment with Explicit Statistical Shape Models in Convolutional Neural Networks." Proceedings of the 10th International Conference on Image Processing Theory, Tools and Applications, IPTA 2020, Virtual, Paris, FRA Institute of Electrical and Electronics Engineers Inc., 2020.
BibTeX: Download