Supervised Descent Method and Its Applications to Face Alignment

By Xuehan Xiong, Fernando De la Torre. 2013

2,077 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

FairFace: Face Attribute Dataset for Balanced Race, Gender, and Age (2019, background): “…image datasets [22, 31, 13, 69, 37, 24, 41, 68, 15, 27, 46, 8, 39] have been proposed and fostered research and development for automated face detection [35, 21], alignment [66, 44], recognition [57, 49], generation [67, 5, 26, 58], modiﬁcation [3, 32, 19], and attribute classiﬁcation [31, 37].”
Deep Learning Face Representation by Joint Identification-Verification (2014, methodology): “We ﬁrst use the recently proposed SDM algorithm [25] to detect 21 facial landmarks.”
The MegaFace Benchmark: 1 Million Faces for Recognition at Scale (2015, methodology): “We further estimate 49 fiducial points and yaw and pitch angles, as computed by the IntraFace3 landmark model [35].”
Face Alignment in Full Pose Range: A 3D Total Solution (2015, background): “Cascaded Linear Regression [11] and traditional nonlinear models [10], [16] are not flexible enough to cover these complex variations in a unified way.”
Weakly-Supervised Evidence Pinpointing and Description (2017, methodology): “{R(k), b} in each iteration is trained with the Supervised Descent Method, the details of which can be found in [14].”
Modeling Stylized Character Expressions via Deep Learning (2016, methodology): “Data Pre-processing For our combined human dataset, Intraface [40] was used to extract 49 facial landmarks.”
High-fidelity Pose and Expression Normalization for face recognition in the wild (2015, methodology): “Benefit from the current breakthrough of face alignment algorithm [47, 16], robustly detecting face landmarks in unconstrained environment has become possible.”
PyMAF: 3D Human Pose and Shape Regression with Pyramidal Mesh Alignment Feedback Loop (2021, methodology): “Figure S4 shows the qualitative differences between each iterative loop in SPIN [9] and Py-MAF, which uses the global features and spatial features for the parameter update respectively.”
Learn to Combine Multiple Hypotheses for Accurate Face Alignment (2013, methodology): “[23] used the cascade regression with SIFT feature, and interpreted the cascade regression procedure from a gradient descent view.”
Extending Long Short-Term Memory for Multi-View Structured Learning (2016, methodology): “The 3 degrees of freedom of a head pose – Pitch, Yaw and Roll – angles were obtained by tracking the child’s face using the IntraFace tracker library [24] and used them as features for the third view.”
Deep Alignment Network: A Convolutional Neural Network for Robust Face Alignment (2017, methodology): “For instance, Supervised Descent Method (SDM) [33] uses SIFT [20] features and a simple linear regressor.”
Mnemonic Descent Method: A Recurrent Process Applied for End-to-End Face Alignment (2016, methodology): “Asthana et al. [3] proposed an incremental learning framework for SDM type methods which supports the total independence of each cascade level.”
OpenFace: An open source facial behavior analysis toolkit (2016, methodology): “For [65] TCDCN CNN [70] EyeTab [63] N/A Intraface SDM [64] ?”
Cost-Effective Active Learning for Deep Image Classification (2017, methodology): “Some earlier works [19], [38] also determined the sample uncertainty referring to a committee of classifiers (i.”
Labeled Faces in the Wild: A Survey (2016, neutral): “Xiong et al., 2013 [107].”