alenaliu / Face-Hallucination-Benchmark

A collection of state-of-the-art face hallucination methods

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Face-Hallucination-Benchmark

A list of face hallucination/face super-resolution resources collected by Junjun Jiang.

Some classical algorithms (including NE, LSR, SR, LcR, LINE, TLcR-RL, and EigTran) that I implemented can be found here.

Classical Patch-based Methods

  • Hallucinating face, FG2000, S. Baker and T. Kanade. [PDF]

  • [NE] Super-resolution through neighbor embedding, CVPR2004, Chang et al. [Web]

  • [LSR] Hallucinating face by position-patch, PR2010, Ma et al. [Web]

  • [SR] Position-patch based face hallucination using convex optimization, SPL2010, Jung et al. [Web]

  • [LcR] Noise robust face hallucination via locality-constrained representation, TMM2104, Jiang et al. [Web]

  • [LINE] Multilayer Locality-Constrained Iterative Neighbor Embedding, TIP2014, Jiang et al. [Web]

  • Face Hallucination Using Linear Models of Coupled Sparse Support, TIP2017, Reuben A. Farrugia et al. [PDF][Web]

  • Hallucinating Face Image by Regularization Models in High-Resolution Feature Space, TIP2018, Jingang Shi et al. [PDF]

  • [TLcR-RL] Context-Patch based Face Hallucination via Thresholding Locality-Constrained Representation and Reproducing Learning, TCYB2018, Junjun Jiang et al. [PDF][Web]

Classical Global Face Methods

  • [EigTran] Hallucinating face by eigentransformation, TSMC-C2005, Xiaogang Wang et al. [Web]

  • Super-resolution of face images using kernel PCA-based prior, TMM2007, A. Chakrabarti et al. [PDF]

  • A Bayesian Approach to Alignment-Based Image Hallucination, ECCV2012, Ce Liu et al. [PDF]

  • A convex approach for image hallucination, AAPRW2013, P. Innerhofer et al. [Code]

  • Structured face hallucination, CVPR2013, Y. Yang et al. [Web]

Classical Two-Step Methods

  • A two-step approach to hallucinating faces: global parametric model and local nonparametric model, CVPR2001, Ce Liu et al. [Web]

  • Hallucinating faces: LPH super-resolution and neighbor reconstruction for residue compensation, PR2007, Yuting Zhuang et al. [PDF]

  • [CCA] Super-resolution of human face image using canonical correlation analysis, PR2010, Hua Huang et al. [PDF]

Deep Learning Method

  • [R-DGN] Ultra-resolving face images by discriminative generative networks, in ECCV2016, Xin Yu et al. [Web]

  • [TDAE] Hallucinating very low-resolution unaligned and noisy face images, CVPR2017, Xin Yu et al. [Web]

  • [TDN] Hallucinating very low-resolution unaligned and noisy face images by transformative discriminative autoencoders, AAAI2017, Xin Yu et al. [Web]

Structure prior based method (componet, landmarks, salience, heatmaps, etc)
  • [CBN] Deep cascaded bi-network for face hallucination, ECCV2016, S. Zhu et al. [PDF][Web]

  • [LCGE] Learning to hallucinate face images via component generation and enhancement, IJCAI2017, Y. Song et al. [PDF][Web]

  • Attention-Aware Face Hallucination via Deep Reinforcement Learning, CVPR2017, Qingxing Cao et al. [PDF][Web]

  • Deep CNN Denoiser and Multi-layer Neighbor Component Embedding for Face Hallucination, IJCAI2018, Junjun Jiang et al. [PDF][Web]

  • [FSRNet] FSRNet: End-to-End learning face super-resolution with facial priors, CVPR, 2018 Yu Chen et al. [PDF][Web]

  • Super-FAN: integrated facial landmark localization and super-resolution of real-world low resolution faces in arbitrary poses with GANs, CVPR2018, Adrian Bulat et al. [PDF][Web]

  • Face super-resolution guided by facial component heatmaps, ECCV2018, Xin Yu et al. [PDF] [Web]

  • A coarse-to-fine face hallucination method by exploiting facial prior knowledge, ICIP2018, Mengyan Li et al. [PDF][Web]

Attribute-Guided method
  • [FaceAttr] Super-resolving very low-resolution face images with supplementary attributes, CVPR2018, Xin Yu et al. [PDF][Web]

  • Attribute-Guided Face Generation Using Conditional CycleGAN, ECCV2018, Yongyi Lu et al. [PDF][Web]

  • Attribute Augmented Convolutional Neural Network for Face Hallucination, CVPRW2018, Cheng-Han Lee et al. [PDF][Web]

Blind face hallucination
  • [GFSRNet] Learning Warped Guidance for Blind Face Restoration, ECCV2018, Xiaoming Li et al. [PDF][Web]

  • To learn image super-resolution, use a GAN to learn how to do image degradation first, ECCV2018, Adrian Bulat et al. [PDF][Web]

Discriminative face hallucination
  • SiGAN: Siamese Generative Adversarial Network for Identity-Preserving Face Hallucination, arXiv2018, Hsu et al. [PDF]

  • Face hallucination using cascaded super-resolution and identity priors, arXiv2018, K. Grm et al. [PDF]

  • Super-Identity Convolutional Neural Network for Face Hallucination, ECCV2018, Kaipeng Zhang et al. [PDF][Web]

  • Verification of Very Low-Resolution Faces Using An Identity-Preserving Deep Face Super-resolution Network, TR2018-116, Esra Ataer-Cansizoglu et al. [PDF]

Image Quality Measurement

  • RMSE, PSNR, SSIM

  • Face recognition rate

  • Mean Opinion Score (MOS)

Databases

Classical databases
Largescale databases

About

A collection of state-of-the-art face hallucination methods