Deep Alignment Network: A convolutional neural network for robust face alignment

6 Jun 2017Marek KowalskiJacek NaruniecTomasz Trzcinski

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... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Face Alignment 300W DAN-Menpo + inter-ocular normalization Mean Error Rate private 3.97 # 5
Face Alignment 300W DAN-Menpo + inter-ocular normalization AUC0.08 private 50.84 # 2
Face Alignment 300W DAN-Menpo + inter-ocular normalization Failure private 1.83% # 2
Face Alignment 300W DAN-Menpo + bounding box diagonal normalization Mean Error Rate private 1.42 # 1