Generating 3D Faces using Convolutional Mesh Autoencoders. A Ranjan, T Bolkart, S Sanyal, MJ Black.
MPI-IS trains CoMA (2018), a mesh autoencoder for generating 3D faces, with 3dMD’s real-world 4D facial data.
UV-GAN: Adversarial Facial UV Map Completion for Pose-invariant Face Recognition. J Deng, S Cheng, N Xue, Y Zhou, S Zafeiriou.
Imperial College London trains UV-GAN (2018) to complete facial textures, using 3dMD’s real captured 3D scans as its core dataset.
Symmetric Shape Morphing for 3D Face and Head Modelling. H Dai, N Pears, W Smith, C Duncan.
University of York and Alder Hey use 3dMD’s real-world Headspace dataset to train SA-CPD, a symmetry-aware shape morphing method for 3D face and head models.
Modelling of Orthogonal Craniofacial Profiles. H Dai, N Pears, C Duncan.
University of York and Alder Hey extend their 2D profile model to three viewpoints using 3dMD’s real-world Headspace dataset, validated in a craniosynostosis outcome study.
Embodied Hands (MANO): Modeling and Capturing Hands and Bodies Together. J Romero, D Tzionas, MJ Black.
MPI-IS build MANO and SMPL+H (2017), articulated hand and body models, with 3dMD’s real-world hand and body data.
Learning a Model of Facial Shape and Expression from 4D Scans. T Li, T Bolkart, MJ Black, H Li, J Romero.
MPI-IS trains FLAME (2017), a facial shape and expression model, with 3dMD’s real-world 4D pose and expression data.
A 3D Morphable Model of Craniofacial Shape and Texture Variation. H Dai, N Pears, W Smith, C Duncan.
University of York and Alder Hey use 3dMD’s real-world Headspace dataset to build the Liverpool-York Head Model (LYHM), the first public 3D morphable model of the full human head.
Symmetry-factored Statistical Modelling of Craniofacial Shape. H Dai, WAP Smith, N Pears.
University of York and Alder Hey use 3dMD’s real-world Headspace dataset to build the first 3D morphable model separating symmetric from asymmetric head shape.
Data-Driven Physics for Human Soft Tissue Animation. M Kim, G Pons-Moll, S Pujades, S Bang, J Kim, MJ Black, SH Lee.
KAIST and MPI-IS build a physics-based body model, VSMPL, with 3dMD’s real-world soft-tissue data from Dyna.
Dynamic FAUST: Registering Human Bodies in Motion. F Bogo, J Romero, G Pons-Moll, MJ Black.
MPI-IS builds Dynamic FAUST (2017), a 4D human motion dataset, with 3dMD’s real-world data, presented at CVPR 2017.