High-fidelity Pose and Expression Normalization for face recognition in the wild

By Xiangyu Zhu, Zhen Lei, Junjie Yan, Dong Yi, Stan Z. Li. Computer Vision and Pattern Recognition, 2015

582 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/f36524cf6442e1649b9763102b7dd40e2afa8c37

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Face Alignment in Full Pose Range: A 3D Total Solution (2015, methodology): “Inspired by the recent achievements in face frontalization [15], [62] which generates the”
Adaptive 3D Face Reconstruction from Unconstrained Photo Collections (2017, methodology): “recently propose an efficient fitting method based only on landmark projection errors [6].”
3D Morphable Face Models—Past, Present, and Future (2019, unknown): “(21) Sometimes the landmarks are allowed to slide on the face surface such that each landmark has a set of vertices to which it could correspond [Zhu et al. 2015].”
Frankenstein: Learning Deep Face Representations Using Small Data (2016, unknown): “05 HPEN [67] 95 .”
Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition (2017, methodology): “As a classic face dataset, it has been used to study face recognition robust to pose [20, 18], illumination [12, 11], and expression [7, 43].”
VITON: An Image-Based Virtual Try-on Network (2017, methodology): “The idea is similar to recent 2D/3D texture warping methods for face synthesis [51, 16], where 2D facial keypoints and 3D pose estimation are utilized for warping.”
Self-Supervised Multi-level Face Model Learning for Monocular Reconstruction at Over 250 Hz (2017, background): “, isolines) [17, 71] or iterate over a fixed vertex set to find 3D contour correspondences [23].”
S^3FD: Single Shot Scale-Invariant Face Detector (2017, background): “Face detection is the key step of many subsequent facerelated applications, such as face alignment [50, 61], face recognition [32, 40, 62], face verification [44, 46] and face tracking [17], etc.”
3D Face Reconstruction by Learning from Synthetic Data (2016, background): “This is sufficient for some tasks, such as in [28], while other require further refinement as applied in [22] for facial reconstruction from multiple images.”
Towards Large-Pose Face Frontalization in the Wild (2017, methodology): “While inpainting methods such as [36] attempt to minimize the impact on quality due to self-occlusions, they still do not retain identity information.”
Beyond Face Rotation: Global and Local Perception GAN for Photorealistic and Identity Preserving Frontal View Synthesis (2017, methodology): “Besides, only the occluded part (right side) of I pred receives the symmetry (a) Proﬁle (b) Ours (c) [33] (d) [41] (e) [8] (f) [44] (g) [12] (h) Frontal loss, i.e. we explicitly pull the right side to be closer to the left.”
Labeled Faces in the Wild: A Survey (2016, methodology): “Finally, two other methods are essentially landmark-based, but used 3D models to fit 3D landmark coordinates to 2D images [108, 114].”
Accurate 3D Face Reconstruction With Weakly-Supervised Learning: From Single Image to Image Set (2019, background): “More recently, model fitting using facial landmarks has gained much popularity with the growth of face alignment techniques [4, 58, 21, 3].”
A Comprehensive Survey on Pose-Invariant Face Recognition (2015, methodology): “To name a few, good surveys exist for illumination-invariant face recognition Zou et al. (2007), 3D face recognition Bowyer et al.”
End-to-End 3D Face Reconstruction with Deep Neural Networks (2017, background): “Among these applications, 3D data has been widely employed in face recognition research to address pose, expression, and illumination variations in facial images, resulting in many publications with state-of-the-art performance [11, 17, 15, 34, 8, 10, 5, 6].”
FaceScape: A Large-Scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Prediction (2020, background): “Following [64], the vertex indices corresponding to contour landmarks of the face are updated after each iteration.”
FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis (2018, methodology): “By following existing works [12, 38, 33], we frontalize all the images in LFW and then evaluate face verification performance on the frontalized images, to verify whether FaceID-GAN retains the identity information properly.”
Disentangled Representation Learning GAN for Pose-Invariant Face Recognition (2017, methodology): “First, some work employ face frontalization [11,14, 31, 40, 43, 45] to synthesize a frontal face, where traditional face recognition methods are applicable.”