• Corpus ID: 238583342

EDFace-Celeb-1M: Benchmarking Face Hallucination with a Million-scale Dataset

@article{Zhang2021EDFaceCeleb1MBF,
  title={EDFace-Celeb-1M: Benchmarking Face Hallucination with a Million-scale Dataset},
  author={Kaihao Zhang and Dongxu Li and Wenhan Luo and Jingyun Liu and Jiankang Deng and Wei Liu and Stefanos Zafeiriou},
  journal={ArXiv},
  year={2021},
  volume={abs/2110.05031}
}
Recent deep face hallucination methods show stunning performance in super-resolving severely degraded facial images, even surpassing human ability. However, these algorithms are mainly evaluated on non-public synthetic datasets. It is thus unclear how these algorithms perform on public face hallucination datasets. Meanwhile, most of the existing datasets do not well consider the distribution of races, which makes face hallucination methods trained on these datasets biased toward some specific… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 66 REFERENCES
Learning Face Hallucination in the Wild
TLDR
A new method of face hallucination is presented, which can consistently improve the resolution of face images even with large appearance variations, based on a novel network architecture called Bi-channel Convolutional Neural Network (Bi-channel CNN).
Super-Identity Convolutional Neural Network for Face Hallucination
TLDR
This paper defines a super-identity loss to measure the identity difference between a hallucinated face and its corresponding high-resolution face within the hypersphere identity metric space and presents a domain-integrated training approach by constructing a robust identity metric for faces from these two domains.
A Comprehensive Survey to Face Hallucination
TLDR
This paper comprehensively surveys the development of face hallucination, including both face super-resolution and face sketch-photo synthesis techniques, and presents a comparative analysis of representative methods and promising future directions.
Copy and Paste GAN: Face Hallucination From Shaded Thumbnails
TLDR
A Copy and Paste Generative Adversarial Network (CPGAN) to recover authentic high-resolution (HR) face images while compensating for low and non-uniform illumination and alleviating the correspondence ambiguity between LR inputs and external HR inputs is proposed.
The MegaFace Benchmark: 1 Million Faces for Recognition at Scale
TLDR
The MegaFace dataset is assembled, both for identification and verification performance, and performance with respect to pose and a persons age is evaluated, as a function of training data size (#photos and #people).
VGGFace2: A Dataset for Recognising Faces across Pose and Age
TLDR
A new large-scale face dataset named VGGFace2 is introduced, which contains 3.31 million images of 9131 subjects, with an average of 362.6 images for each subject, and the automated and manual filtering stages to ensure a high accuracy for the images of each identity are described.
Generalized Face Super-Resolution
  • K. Jia, S. Gong
  • Mathematics, Medicine
    IEEE Transactions on Image Processing
  • 2008
TLDR
An automatic face alignment algorithm capable of pixel-wise alignment by iteratively warping the probing face to its projection in the space of training face images and novelty of the approach in coping with multimodal hallucination and its robustness in automatic alignment under practical imaging conditions are presented.
Hallucinating Compressed Face Images
A face hallucination algorithm is proposed to generate high-resolution images from JPEG compressed low-resolution inputs by decomposing a deblocked face image into structural regions such as facial
Deep Cascaded Bi-Network for Face Hallucination
TLDR
A novel framework for hallucinating faces of unconstrained poses and with very low resolution, which allows exceptional hallucination quality on in-the-wild low-res faces with significant pose and illumination variations is presented.
Face Super-Resolution Guided by Facial Component Heatmaps
TLDR
This paper proposes a method that explicitly incorporates structural information of faces into the face super-resolution process by using a multi-task convolutional neural network (CNN) and achieves superior face hallucination results and outperforms the state-of-the-art.
...
1
2
3
4
5
...