The Digital Emily project: photoreal facial modeling and animation

By Oleg Alexander, Mike Rogers, W. Lambeth, M. Chiang, Paul E. Debevec. 2009

142 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, 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 Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

A Hybrid Model for Identity Obfuscation by Face Replacement (2018, methodology): “ parameters are the coecients of the PCA vectors constructed from 200 high-quality face scans [10]. are the coecients of the expression basis vectors computed using PCA on selected blend shapes of [29] A Hybrid Model for Identity Obfuscation by Face Replacement 5 and [30]. We use 80 , 80 and 64 para”
Example-based facial rigging (2010, methodology): “Building such a linear facial rig for highly realistic animation was recently demonstrated by [Alexander et al. 2009], though each blendshape was still hand-crafted by animators.”
3D Virtual worlds and the metaverse: Current status and future possibilities (2013, background): “The company’s Faceware product has seen effective applications in a variety of computer games and animated films as well as research that specifically targets the uncanny valley phenomenon [Alexander et al. 2009].”
MoFA: Model-Based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstruction (2017, methodology): “We obtain the expression basis by applying PCA to the combined set of blendshapes of [1] and [6], which have been re-targeted to the face topology of [3] using deformation transfer [49].”
Total Moving Face Reconstruction (2014, background): “Structured light [45] and light stages [16,2,3,25] provide the ability to use multiple synchronized and calibrated lights for reconstruction.”
Capture, Learning, and Synthesis of 3D Speaking Styles (2019, background): “Alexander et al. [3], Wu et al. [61], and Laine et al. [35] build subject-speciﬁc face-rigs from high-resolution face scans and animate these rigs with video-based animation systems.”
Realtime facial animation with on-the-fly correctives (2013, methodology): “Detailed 3D facial tracking from video is generally achieved at the cost of a controlled environment, expensive computation, and the use of a care-fully crafted 3D tracking model [Borshukov et al. 2005; Alexander et al. 2009; Vlasic et al. 2005].”
StyleRig: Rigging StyleGAN for 3D Control Over Portrait Images (2020, methodology): “We obtain this model by applying PCA to a set of blendshapes [2, 6] which have been transferred to the topology of the shape and reflectance models.”
Reconstruction of Personalized 3D Face Rigs from Monocular Video (2016, methodology): “…cover facial expressions by adding Ke = 75 delta blendshapes (i.e., displacements from the rest pose) taken from a combination of the Emily model [Alexander et al. 2009] and the FaceWarehouse database [Cao et al. 2014b]: Expression: Pe(α, δ) = P s(α) + Ee eδ, (4) where the matrix Ee ∈ R3N×Ke…”
Learning a model of facial shape and expression from 4D scans (2017, background): “Publication date: November 2017. that individual and/or involve significant input from a 3D artist (e.g. [Alexander et al. 2009]).”
A Survey on Human Motion Analysis from Depth Data (2013, background): “A typical application is performance-based facial animation where the non-rigid surface of the face is tracked and the motion is transferred to a virtual face [98, 99].”
High resolution passive facial performance capture (2010, methodology): “Recently, Alexander et al. [2009] created a photoreal facial modeling and animation system in the Digital Emily Project.”
Real-time expression transfer for facial reenactment (2015, methodology): “…3D multi-linear model from example data spanning a large variety of identities and expressions [Vlasic et al. 2005], or by purposely building a person-specific blend shape model from scans of an actor using specialized hardware [Eisert and Girod 1998; Alexander et al. 2009; Weise et al. 2011].”
Production-level facial performance capture using deep convolutional neural networks (2016, methodology): “…reconstruction of individual input frames [Beeler et al. 2011; Bradley et al. 2010; Furukawa and Ponce 2009; Fyffe et al. 2011; Shi et al. 2014; Valgaerts et al. 2012] that often require carefully crafted 3D tracking model [Alexander et al. 2009; Borshukov et al. 2005; Vlasic et al. 2005].”
High-quality passive facial performance capture using anchor frames (2011, unknown): “Finally, combining markers and structured light with a light stage has proven to yield impressive facial performance capture results [Ma et al. 2008; Alexander et al. 2009].”
Self-Supervised Multi-level Face Model Learning for Monocular Reconstruction at Over 250 Hz (2017, methodology): “The subspace of expression variations is spanned by the vectors {bgk} ms+me k=ms+1 which were created using PCA from a subset of blendshapes of [2] and [18].”