Deep Alignment Network: A Convolutional Neural Network for Robust Face Alignment

By Marek Kowalski, Jacek Naruniec, Tomasz Trzcinski. 2017 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2017

334 citations recorded by the source corpus.

In this paper, we propose Deep Alignment Network (DAN), a robust face alignment method based on a deep neural network architecture. DAN consists of multiple stages, where each stage improves the locations of the facial landmarks estimated by the previous stage. Our method uses entire face images at all stages, contrary to the recently proposed face alignment methods that rely on local patches. This is possible thanks to the use of landmark heatmaps which provide visual information about landmark locations estimated at the previous stages of the algorithm. The use of entire face images rather than patches allows DAN to handle face images with large variation in head pose and difficult initializations. An extensive evaluation on two publicly available datasets shows that DAN reduces the state-of-the-art failure rate by up to 70%. Our method has also been submitted for evaluation as part of the Menpo challenge.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/98b83d56ebc4714dcb6b72bfdb426d739b7fb86f

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression (2019, background): “Heatmap regression , which regresses a heatmap gen-Foregound 2d), the heatmap becomes much sharper on landmarks. erated from landmark coordinates, is widely used for face alignment [5, 29, 64, 50].”
Whole-Body Human Pose Estimation in the Wild (2020, neutral): “DAN [27] - 3.”
FSRNet: End-to-End Learning Face Super-Resolution with Facial Priors (2017, methodology): “Despite the disadvantage in the input image resolution, our method outperforms most recent methods and is competitive with the state of the art [19].”
Deep High-Resolution Representation Learning for Visual Recognition (2019, methodology): “The estimated segmentation maps are improved by combining the ﬁne segmentation score maps estimated from intermediate low-level medium-resolution representations [99], or iterating the processes [76].”
The face of art (2019, methodology): “They annotated 1446 images out of the 109Manga dataset [Ogawa et al. 2018] for training a CNN based on the Deep Alignment Network (DAN) architecture [Kowalski et al. 2017].”
Neural best-buddies (2018, methodology): “Most of the existing works on part localization or keypoint prediction focus on either facial landmark localization [Belhumeur et al. 2013; Kowalski et al. 2017] or human pose estimation [Gkioxari et al.”