Large-Pose Face Alignment via CNN-Based Dense 3D Model Fitting

By Amin Jourabloo, Xiaoming Liu. Computer Vision and Pattern Recognition, 2016

317 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/262cb6f0b9c910792b23aa4c3371652a976c01d1

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

Large Pose 3D Face Reconstruction from a Single Image via Direct Volumetric CNN Regression (2017, methodology): “• Because of this fundamental difference, our method is also radically different in terms of the CNN architecture used: we used one that is able to make spatial predictions at a voxel level, as opposed to the networks of [28, 9] which holistically predict the 3DMM parameters.”
Adaptive 3D Face Reconstruction from Unconstrained Photo Collections (2017, background): “Hopefully the improved capability of landmark alignment for large-pose faces [20, 21, 42] will further improve 3D reconstruction performance.”
How Far are We from Solving the 2D & 3D Face Alignment Problem? (and a Dataset of 230,000 3D Facial Landmarks) (2017, methodology): “Nevertheless, [4] is evaluated on [20] which is a relatively small dataset (3900 images for training and 1200 for testing) and [6] on [19] which is of moderate size (16,2000 images for training and 4,900 for testing), includes mainly images collected in the lab and does not cover the full spectrum o”
Extreme 3D Face Reconstruction: Looking Past Occlusions (2017, background): “Detected facial landmarks can be used to constrain 3D face shapes [24, 53].”
Nonlinear 3D Face Morphable Model (2018, methodology): “Hence, there is increasing attention in conducting face alignment by fitting a 3D face model to a single 2D image [16–19, 21, 24, 45].”
Regressing Robust and Discriminative 3D Morphable Models with a Very Deep Neural Network (2016, unknown): “, [19, 36]) and the CNN based approaches of [20, 47].”
Unrestricted Facial Geometry Reconstruction Using Image-to-Image Translation (2017, methodology): “While using a neural network for face reconstruction was proposed in the past [33, 34, 43, 48, 24], previous methods were still limited by the expressiveness of the linear model.”
Extreme 3D Face Reconstruction: Seeing Through Occlusions (2017, background): “Detected facial landmarks can be used to constrain 3D face shapes [25, 58].”
RetinaFace: Single-stage Dense Face Localisation in the Wild (2019, methodology): “8(a) and 8(b), we give the CED curves of state-of-the-art methods [12, 70, 75, 25, 3] as well as RetinaFace.”
Multi-Task Convolutional Neural Network for Pose-Invariant Face Recognition (2017, unknown): “We use PIFA [17] to estimate 34 landmarks and the yaw angle, which deﬁnes three groups: right proﬁle [ − 90 ◦ , − 30 ◦ ), frontal [ − 30 ◦ , 30 ◦ ], and left proﬁle ( 30 ◦ , 90 ◦ ].”
Face anti-spoofing using patch and depth-based CNNs (2017, methodology): “However, compared to other face related problems, such as face recognition [27, 43] and face alignment [20], there are still substantially less efforts and exploration on face anti-spoofing using deep learning techniques.”
Towards Fast, Accurate and Stable 3D Dense Face Alignment (2020, background): “Recent studies are mainly divided into two categories: 3D Morphable Model (3DMM) parameters regression [29,59,34,36,48,60,23] and dense vertices regression [28,18].”
Wing Loss for Robust Facial Landmark Localisation with Convolutional Neural Networks (2017, background): “The second strategy, which has become very popular in recent years, is to use 3D face models [76, 28, 2, 38, 29].”
End-to-End 3D Face Reconstruction with Deep Neural Networks (2017, methodology): “Compared with [33, 11, 25], one major difference of our deep neural network is that it is end-to-end and takes only a single RGB image as input.”
Deep Kinematic Pose Regression (2016, methodology): “, eyes or nose on human face [16], and wings or tail on bird [38].”
Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network (2018, methodology): “AFLW-LFPA is another extension of AFLW dataset constructed by [10].”
Facial Landmark Detection: A Literature Survey (2018, methodology): “Similarly, in [50], in a cascaded manner, the whole facial appearance is used in the ﬁrst cascaded CNN model to predict the updates of 3D shape parameters and pose, while the local patches are used in the later cascaded CNN models to reﬁne the landmarks.”
3D Face Morphable Models "In-the-Wild" (2017, methodology): “That is, starting with the 3D fits provided by using only the 2D landmarks [20], a texture model can be learned using the above procedure.”