Corpus ID: 199000983

Dressing 3D Humans using a Conditional Mesh-VAE-GAN

@article{Ma2019Dressing3H,
  title={Dressing 3D Humans using a Conditional Mesh-VAE-GAN},
  author={Qianli Ma and Siyu Tang and Sergi Pujades and Gerard Pons-Moll and Anurag Ranjan and Michael J. Black},
  journal={ArXiv},
  year={2019},
  volume={abs/1907.13615}
}
Three-dimensional human body models are widely used in the analysis of human pose and motion. [...] Key Method Specifically, we train a conditional Mesh-VAE-GAN to learn the clothing deformation from the SMPL body model, making clothing an additional term on SMPL. Our model is conditioned on both pose and clothing type, giving the ability to draw samples of clothing to dress different body shapes in a variety of styles and poses.Expand

Topics from this paper

Building High-Fidelity Human Body Models From User-Generated Data
TLDR
The proposed approach outperforms the state-of-the-arts in the accuracy of both human body shape and pose estimation and can build believable and animatable human body models robustly based on low-resolution single images taken by smartphones. Expand
Human Motion Tracking with Less Constraint of Initial Posture from a Single RGB-D Sensor
TLDR
A human motion capture system combined with human priors and performance capture that only uses a single RGB-D sensor that can be used more comfortably while maintaining credible geometric warping and remaining free of self-scanning constraints is proposed. Expand
UCCTGAN: Unsupervised Clothing Color Transformation Generative Adversarial Network
TLDR
Experimental results show that the proposed Unsupervised Clothing Color Transformation Generative Adversarial Network (UCCTGAN) has the ability to generate convincing color transformation results. Expand
Deriving the configuration space of a rectangular cloth
Forty years ago the notion of configuration space (C–space) revolutionised robot motion planning for rigid and articulated objects. Despite great progress, handling deformable materials has remainedExpand
Towards Geometric Understanding of Motion
TLDR
This work introduces Competitive Collaboration, where several neural networks are constrained by geometry and can jointly learn about structure and motion in the scene without any labels and shows that jointly learning single view depth prediction, camera motion, optical flow and motion segmentation using competitive Collaboration achieves state-of-the-art results among unsupervised approaches. Expand
CLOTH3D: Clothed 3D Humans
TLDR
A Conditional Variational Auto-Encoder based on graph convolutions (GCVAE) to learn garment latent spaces allows for realistic generation of 3D garments on top of SMPL model for any pose and shape. Expand
Unsupervised Geometric Disentanglement for Surfaces via CFAN-VAE
TLDR
This work introduces a novel mesh feature, the conformal factor and normal feature (CFAN), for use in mesh convolutional autoencoders and proposes CFAN-VAE, a novel architecture that disentangles identity and pose using the CFAN feature and parallel transport convolution. Expand

References

SHOWING 1-10 OF 80 REFERENCES
SMPL: a skinned multi-person linear model
TLDR
The Skinned Multi-Person Linear model (SMPL) is a skinned vertex-based model that accurately represents a wide variety of body shapes in natural human poses that is compatible with existing graphics pipelines and iscompatible with existing rendering engines. Expand
DRAPE: DRessing Any PErson
TLDR
An algorithm that dresses the body with a garment that is customized to fit and possesses realistic wrinkles is described that is significantly more efficient than physical simulation. Expand
Detailed Human Shape Estimation From a Single Image by Hierarchical Mesh Deformation
TLDR
A novel learning-based framework that combines the robustness of parametric model with the flexibility of free-form 3D deformation and is able to restore detailed human body shapes beyond skinned models is proposed. Expand
Detailed, Accurate, Human Shape Estimation from Clothed 3D Scan Sequences
TLDR
This work contributes a new approach to recover a personalized shape of the person by estimating body shape under clothing from a sequence of 3D scans, which outperforms the state of the art in both pose estimation and shape estimation, qualitatively and quantitatively. Expand
Analyzing Clothing Layer Deformation Statistics of 3D Human Motions
TLDR
It is shown that this model not only allows to reproduce previous retargeting works, but generalizes the data generation capabilities to other semantic parameters such as clothing variation and size, or physical material parameters with synthetically generated training sequence, paving the way for many kinds of capture data-driven creation and augmentation applications. Expand
DeepWrinkles: Accurate and Realistic Clothing Modeling
TLDR
An entirely data-driven approach to realistic cloth wrinkle generation is claimed, which leads to unprecedented high-quality rendering of clothing deformation sequences, where fine wrinkles from (real) high resolution observations can be recovered. Expand
The Naked Truth: Estimating Body Shape Under Clothing
TLDR
Results on a novel database of thousands of images of clothed and "naked" subjects, as well as sequences from the HumanEva dataset, suggest the method may be accurate enough for biometric shape analysis in video. Expand
A Layered Model of Human Body and Garment Deformation
TLDR
The proposed deformation model provides intuitive control over the three parameters independently, while producing aesthetically pleasing deformations of both the garment and the human body. Expand
Video Based Reconstruction of 3D People Models
TLDR
This paper describes a method to obtain accurate 3D body models and texture of arbitrary people from a single, monocular video in which a person is moving and presents a robust processing pipeline to infer 3D model shapes including clothed people with 4.5mm reconstruction accuracy. Expand
Generating 3D faces using Convolutional Mesh Autoencoders
TLDR
This work introduces a versatile model that learns a non-linear representation of a face using spectral convolutions on a mesh surface and shows that, replacing the expression space of an existing state-of-the-art face model with this model, achieves a lower reconstruction error. Expand
...
1
2
3
4
5
...