Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks

@article{Feng2018WingLF,
  title={Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks},
  author={Zhenhua Feng and Josef Kittler and Muhammad Awais and P. Huber and Xiaojun Wu},
  journal={2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition},
  year={2018},
  pages={2235-2245}
}
We present a new loss function, namely Wing loss, for robust facial landmark localisation with Convolutional Neural Networks (CNNs. [] Key Method The new loss amplifies the impact of errors from the interval (-w, w) by switching from L1 loss to a modified logarithm function. To address the problem of under-representation of samples with large out-of-plane head rotations in the training set, we propose a simple but effective boosting strategy, referred to as pose-based data balancing.

Figures and Tables from this paper

Rectified Wing Loss for Efficient and Robust Facial Landmark Localisation with Convolutional Neural Networks
TLDR
This paper presents a new loss function, namely Rectified Wing (RWing) loss, for regression-based facial landmark localisation with Convolutional Neural Networks (CNNs), and proposes a simple but effective boosting strategy, referred to as pose-based data balancing, for under-representation of samples with large pose variations.
Residual Neural Network and Wing Loss for Face Alignment Network
  • Li Wang, Wei Xiang
  • Computer Science
    2019 IEEE 14th International Conference on Intelligent Systems and Knowledge Engineering (ISKE)
  • 2019
TLDR
A simple but effective data enhancement method is proposed to deal with the problem of unbalanced data processing by randomly rotating the training samples, amplifying and the like, and the experimental results obtained by the method on the 300W dataset indicate the advantages of this method.
2D Wasserstein Loss for Robust Facial Landmark Detection
Adaptive Wing Loss for Robust Face Alignment via Heatmap Regression
TLDR
A novel loss function is proposed, named Adaptive Wing loss, that is able to adapt its shape to different types of ground truth heatmap pixels, that penalizes loss more on foreground pixels while less on background pixels.
Stacked Multi-Target Network for Robust Facial Landmark Localisation
TLDR
This work thoroughly analyse regression-based face alignment methods and introduces a novel stacked multi-target network for robust facial landmark localisation, which performs superiorly to the state-of-the-art approaches.
Mining Hard Augmented Samples for Robust Facial Landmark Localization With CNNs
TLDR
Different data augmentation techniques that can be used to generate sufficient data for training CNN-based facial landmark localization systems are investigated and an online hard augmented example mining (HAEM) strategy is advocated for further performance boosting.
Efficient Facial Landmark Localization Based on Binarized Neural Networks
TLDR
This paper introduces a new network architecture to calculate the binarized models, referred to as Amplitude Convolutional Networks (ACNs), based on the proposed asynchronous back propagation algorithm, and reduces the storage space of convolutional filters by a factor of 32 compared with the full-precision models on dataset LFW+Webface, CelebA, BioID and 300W.
Separable Batch Normalization for Robust Facial Landmark Localization with Cross-protocol Network Training
TLDR
A novel Separable Batch Normalization module with a Cross-protocol Network Training (CNT) strategy for robust facial landmark localization and a novel attention mechanism that assigns different weights to each branch for automatic selection in an effective style are presented.
Subpixel Heatmap Regression for Facial Landmark Localization
TLDR
This work proposes a new approach for the heatmap encoding and decoding process by leveraging the underlying continuous distribution and introduces a Siamese-based training that enforces heatmap consistency across various geometric image transformations.
A Robust Facial Landmark Detector with Mixed Loss
TLDR
This work proposes a robust facial landmark detection algorithm based on coordinates regression in an end-to-end training fashion that achieves real-time speed on 300-W with 68 landmarks, which runs at 85 FPS on a Tesla v100 GPU.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 78 REFERENCES
Stacked Hourglass Network for Robust Facial Landmark Localisation
TLDR
A Stacked Hourglass Network for robust facial landmark localisation by adopting a supervised face transformation to remove the translation, scale and rotation variation of each face, and employing a deep convolutional neural network to increase the capacity of the regression model.
Facial Landmark Detection with Tweaked Convolutional Neural Networks
TLDR
A novel CNN architecture, specialized to regress the facial landmark coordinates of faces in specific poses and appearances is described, shown to outperform existing landmark detection methods in an extensive battery of tests on the AFW, ALFW, and 300W benchmarks.
Unconstrained Facial Landmark Localization with Backbone-Branches Fully-Convolutional Networks
TLDR
A novel Backbone-Branches Fully-Convolutional Neural Network (BB-FCN), which produces facial landmark response maps directly from raw images without relying on pre-process or sliding window approaches, achieving superior performances over other state-of-the-arts under both the constrained and in the wild scenarios.
Deep Convolutional Network Cascade for Facial Point Detection
TLDR
The proposed approach outperforms state-of-the-art methods in both detection accuracy and reliability and can avoid local minimum caused by ambiguity and data corruption in difficult image samples due to occlusions, large pose variations, and extreme lightings.
Binarized Convolutional Landmark Localizers for Human Pose Estimation and Face Alignment with Limited Resources
TLDR
The first to study the effect of neural network binarization on localization tasks, namely human pose estimation and face alignment, and proposes a novel hierarchical, parallel and multi-scale residual architecture that yields large performance improvement over the standard bottleneck block while having the same number of parameters.
Pose-Invariant Face Alignment with a Single CNN
TLDR
A new layer, named visualization layer, is proposed, which can be integrated into the CNN architecture and enables joint optimization with different loss functions and demonstrates state-of-the-art accuracy, while reducing the training time by more than half compared to the typical cascade of CNNs.
Facial Landmark Detection by Deep Multi-task Learning
TLDR
A novel tasks-constrained deep model is formulated, with task-wise early stopping to facilitate learning convergence and reduces model complexity drastically compared to the state-of-the-art method based on cascaded deep model.
Dense Face Alignment
TLDR
Experimental results show the method not only provides high-quality, dense 3D face fitting but also outperforms the state-of-the-art facial landmark detection methods on challenging datasets.
Joint Multi-View Face Alignment in the Wild
TLDR
This paper proposes the first, to the best of the knowledge, joint multi-view convolutional network to handle large pose variations across faces in-the-wild, and elegantly bridge face detection and facial landmark localization tasks.
Reliable Crowdsourcing and Deep Locality-Preserving Learning for Expression Recognition in the Wild
TLDR
A new DLP-CNN (Deep Locality-Preserving CNN) method, which aims to enhance the discriminative power of deep features by preserving the locality closeness while maximizing the inter-class scatters, is proposed.
...
1
2
3
4
5
...