Face Synthesis for Eyeglass-Robust Face Recognition

@article{Guo2018FaceSF,
  title={Face Synthesis for Eyeglass-Robust Face Recognition},
  author={Jianzhu Guo and Xiangyu Zhu and Zhen Lei and S. Li},
  journal={ArXiv},
  year={2018},
  volume={abs/1806.01196}
}
In the application of face recognition, eyeglasses could significantly degrade the recognition accuracy. A feasible method is to collect large-scale face images with eyeglasses for training deep learning methods. However, it is difficult to collect the images with and without glasses of the same identity, so that it is difficult to optimize the intra-variations caused by eyeglasses. In this paper, we propose to address this problem in a virtual synthesis manner. The high-fidelity face images… 
ByeGlassesGAN: Identity Preserving Eyeglasses Removal for Face Images
TLDR
A novel image-to-image GAN framework for eyeglass removal, called ByeGlassesGAN, which is used to automatically detect the position of eyeglasses and then remove them from face images, and demonstrates significant improvement in face recognition accuracy for face images with glasses.
Unsupervised Eyeglasses Removal in the Wild
TLDR
A unified eyeglass removal model called the eyeglasses removal generative adversarial network (ERGAN) is proposed, which could handle different types of glasses in the wild and achieves a competitive removal quality in terms of realism and diversity.
FACE VERIFICATION USING CONVOLUTIONAL NEURAL NETWORK WITH PARTIAL TRIPLET LOSS ON FACE WEARING GLASSES
TLDR
CNN algorithm with siamese architecture is used to extract features from the face and proposed to use the partial triplet loss function to partially minimize the optimization problem on the network to handle partial occlusion due to face wearing glasses.
Portrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data
TLDR
This paper presents a new synthetic portrait dataset with both intermediate and final supervisions for both the detection and removal tasks, and proposes a novel framework to remove eyeglasses as well as their cast shadows from face images simultaneously.
Improving Face Anti-Spoofing by 3D Virtual Synthesis
TLDR
A method to synthesize virtual spoof data in 3D space to alleviate the problem of expensive spoof data acquisition and open up new possibilities for advancing face anti-spoofing using cheap and large-scale synthetic data.
On Improving the Generalization of Face Recognition in the Presence of Occlusions
TLDR
O improved the generalization ability of face recognition under occlusions by 10.17% in a single-image-based setting and outperformed the baseline by approximately 2% in terms of rank-1 accuracy in an image-set-based scenario.
Cross-Quality LFW: A Database for Analyzing Cross- Resolution Image Face Recognition in Unconstrained Environments
TLDR
This work proposes a new standardized benchmark dataset and evaluation protocol derived from the famous Labeled Faces in the Wild (LFW), and shows that these models perform differently in cross-quality cases, and hence the generalizing capability is not accurately predicted by their performance on LFW.
3DMA: A Multi-modality 3D Mask Face Anti-spoofing Database
TLDR
A multi-modality 3D mask face anti-spoofing database named 3DMA, which contains 920 videos of 67 genuine subjects wearing 48 kinds of 3D masks, captured in visual (VIS) and near-infrared (NIR) modalities, is released.
Out-of-Distribution Detection for Reliable Face Recognition
TLDR
This paper proposes to detect out-of-distribution samples based on uncertainty prediction and the L2-norm of features, so as to effectively filter out non-face and low-quality faces.
...
...

References

SHOWING 1-10 OF 31 REFERENCES
Learning sparse feature for eyeglasses problem in face recognition
TLDR
The proposed method applies the sparse representation technique in a local feature space so as to be more tolerant to mis-alignment and abnormal specular pixel values, which leads to a good balance between efficiency and robustness.
Automatic eyeglasses removal from face images
TLDR
An intelligent image editing and face synthesis system that automatically removes eyeglasses from an input frontal face image by adopting a statistical analysis and synthesis approach to learn the mapping between pairs of face images with and without eyeglass from a database.
Eyeglasses removal from facial images
A Lightened CNN for Deep Face Representation
TLDR
A lightened CNN framework to learn a compact embedding for face representation by introducing the concept of maxout in the fully connected layer to the convolution layer, which leads to a new activation function, named Max-Feature-Map (MFM).
DeepFace: Closing the Gap to Human-Level Performance in Face Verification
TLDR
This work revisits both the alignment step and the representation step by employing explicit 3D face modeling in order to apply a piecewise affine transformation, and derive a face representation from a nine-layer deep neural network.
Deep Learning Face Representation by Joint Identification-Verification
TLDR
This paper shows that the face identification-verification task can be well solved with deep learning and using both face identification and verification signals as supervision, and the error rate has been significantly reduced.
Glasses removal from facial image using recursive error compensation
TLDR
The experimental results show that the proposed method provides an effective solution to the problem of glasses occlusion and it is believed that this method can also be used to enhance the performance of face recognition systems.
Face recognition in unconstrained videos with matched background similarity
TLDR
A comprehensive database of labeled videos of faces in challenging, uncontrolled conditions, the ‘YouTube Faces’ database, along with benchmark, pair-matching tests are presented and a novel set-to-set similarity measure, the Matched Background Similarity (MBGS), is described.
Magic Glasses: From 2D to 3D
TLDR
This paper proposes an eyeglasses segmentation method using a convolutional neural network-based parsing algorithm to label the glasses pixels, followed with a proposed symmetry-based level-set optimization algorithm to refine the contour of the eyeglass, and proposes a silhouette-based 3D deformation method to deform a 3D eyeg glasses model selected from a predefined 3D model database.
SphereFace: Deep Hypersphere Embedding for Face Recognition
TLDR
This paper proposes the angular softmax (A-Softmax) loss that enables convolutional neural networks (CNNs) to learn angularly discriminative features in deep face recognition (FR) problem under open-set protocol.
...
...