Deep Alignment Network: A Convolutional Neural Network for Robust Face Alignment
Marek Kowalski,Jacek Naruniec,Tomasz Trzcinski +2 more
- pp 2034-2043
Reads0
Chats0
TLDR
The use of entire face images rather than patches allows DAN to handle face images with large variation in head pose and difficult initializations, and reduces the state-of-the-art failure rate by up to 70%.Abstract:
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.read more
Citations
More filters
Book ChapterDOI
Research on Monitoring Methods for Electricity Hall Staff Based on Autonomous Updating and Semi-supervising Model
TL;DR: A self-updating and semi-supervised model used to optimize the existing monitoring system and can meet the requirements of high timeliness for personnel on-the-job monitoring, and can also identify the appointed staff on the basis of facial features mining data in batches.
L abel e ncoding for r egression n etworks
TL;DR: BEL is introduced, which generalizes the application of binary classification to regression by providing a framework for considering arbitrary multi-bit values when encoding target values and achieves state-of-the-art accuracies for several regression benchmarks.
Posted Content
Subpixel Heatmap Regression for Facial Landmark Localization
TL;DR: In this article, a Siamese-based training was proposed to enforce heatmap consistency across various geometric image transformations, which achieved state-of-the-art results in facial landmark localization.
Proceedings ArticleDOI
ABN: Anti-Blur Neural Networks for Multi-Stage Deformable Image Registration
TL;DR: Zhang et al. as discussed by the authors used a pair of short-term registration and long-term memory networks to learn the nonlinear deformations at each stage and combined all the previous deformations to allow an interpolation to perform on the raw image directly and preserve image sharpness.
Proceedings ArticleDOI
RULe: Relocalization-Uniformization-Landmark Estimation Network for Real-Time Face Alignment in Degraded Conditions
TL;DR: In this paper , the authors propose a three-staged Relocalization-Uniformization-Landmark Estimation network to estimate the position of salient landmarks on face images or videos.
References
More filters
Proceedings Article
Adam: A Method for Stochastic Optimization
Diederik P. Kingma,Jimmy Ba +1 more
TL;DR: This work introduces Adam, an algorithm for first-order gradient-based optimization of stochastic objective functions, based on adaptive estimates of lower-order moments, and provides a regret bound on the convergence rate that is comparable to the best known results under the online convex optimization framework.
Proceedings Article
Very Deep Convolutional Networks for Large-Scale Image Recognition
Karen Simonyan,Andrew Zisserman +1 more
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers.
Journal ArticleDOI
Distinctive Image Features from Scale-Invariant Keypoints
TL;DR: This paper presents a method for extracting distinctive invariant features from images that can be used to perform reliable matching between different views of an object or scene and can robustly identify objects among clutter and occlusion while achieving near real-time performance.
Journal Article
Dropout: a simple way to prevent neural networks from overfitting
TL;DR: It is shown that dropout improves the performance of neural networks on supervised learning tasks in vision, speech recognition, document classification and computational biology, obtaining state-of-the-art results on many benchmark data sets.
Proceedings ArticleDOI
Histograms of oriented gradients for human detection
Navneet Dalal,Bill Triggs +1 more
TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.