We add the homepage of our dataset, you can visit here.
The EDFace-Celeb dataset is released under CC BY-NC-ND license.
We construct a large-scale dataset for face analysis. It consists of three sub-datasets, i.e., EDFace-Celeb-1M, EDFace-Celeb-150K and EDFace-Celeb-Real datasets. The EDFace-Celeb-1M and EDFace-Celeb-150K datasets provide different settings for face super-resolution (FH128, FH512), face hallucination (FH128, FH512) and blind face restoration (BFR128, BFR512).
- DIC: Results
- DIC-GAN: Results
- HiFaceGAN: Results
- Wavelet: Results
- EDSR: Results
- RCAN: Results
- RDN: Results
- HAN: Results
EDFace-Celeb-1M (BFR128) : Blind Face Restoration (HQ, LQ: Blur, JPEG artifact, Noise, SR, Full, Full_X2, Full_X4, Full_X8)
- EDFace-Celeb-1M: Benchmarking Face Hallucination with a Million-scale Dataset (TPAMI2022)
- Blind Face Restoration: Benchmark Datasets and a Baseline Model
EDFace-Celeb-150K (BFR512): Blind Face Restoration (HQ, LQ: Blur, JPEG artifact, Noise, SR, Full, Full_X2, Full_X4, Full_X8)
- EDFace-Celeb-1M: Benchmarking Face Hallucination with a Million-scale Dataset (TPAMI2022)
- Blind Face Restoration: Benchmark Datasets and a Baseline Model
If you think the EDFace-Celeb dataset is useful for your research, please cite the following paper.
@inproceedings{zhang2022edface,
title={EDFace-Celeb-1M: Benchmarking Face Hallucination with a Million-scale Dataset},
author={Zhang, Kaihao and Li, Dongxu and Luo, Wenhan and Liu, Jingyu and Deng, Jiankang and Liu, Wei and Stefanos Zafeiriou},
booktitle={IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)},
year={2022}
}