Posture-invariant statistical shape analysis using Laplace operator

Wuhrer, Stefanie; Shu, Chang; Xi, Pengcheng

doi:10.1016/j.cag.2012.03.026

Cited by 22 publications

(22 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If our goal is to understand the variability of human shapes, we must then factor out the variability due to pose variations among the subjects. Various approaches have been tried towards this end, including PCA in a Riemannian symmetric space [Fletcher et al 2004], multivariate tensor-based morphometry using holomorphic forms [Wang et al 2010], tensor ICA [Vasilescu and Terzopoulos 2007], the use of Laplace coordinates for points [Wuhrer et al 2012], and others [Nain et al 2007]. Only recently has an approach been proposed in these communities for comparing shapes intrinsically [Lai et al 2010] using a spectral L 2 distance, but the approach suffers from the usual sign ambiguities (or more generally rotations within an eigenspace) of the eigenfunctions in spectral embeddings.…”

Section: Related Workmentioning

confidence: 99%

Map-based exploration of intrinsic shape differences and variability

et al. 2013

View full text Add to dashboard Cite

Figure 1: The notion of shape difference defined in this paper provides a way to compare deformations between shape pairs. This allows us to recognize similar expressions of shape A (top row) to those of shape B (bottom row), without correspondences between A and B and without any prior learning process. AbstractWe develop a novel formulation for the notion of shape differences, aimed at providing detailed information about the location and nature of the differences or distortions between the two shapes being compared. Our difference operator, derived from a shape map, is much more informative than just a scalar global shape similarity score, rendering it useful in a variety of applications where more refined shape comparisons are necessary. The approach is intrinsic and is based on a linear algebraic framework, allowing the use of many common linear algebra tools (e.g, SVD, PCA) for studying a matrix representation of the operator. Remarkably, the formulation allows us not only to localize shape differences on the shapes involved, but also to compare shape differences across pairs of shapes, and to analyze the variability in entire shape collections based on the differences between the shapes. Moreover, while we use a map or correspondence to define each shape difference, consistent correspondences between the shapes are not necessary for comparing shape differences, although they can be exploited if available. We give a number of applications of shape differences, including parameterizing the intrinsic variability in a shape collection, exploring shape collections using local variability at different scales, performing shape analogies, and aligning shape collections.

show abstract

Section: Related Workmentioning

confidence: 99%

Map-based exploration of intrinsic shape differences and variability

et al. 2013

View full text Add to dashboard Cite

show abstract

“…While this approach leads to better shape spaces, it is difficult to directly apply this approach to the S-SCAPE spaces learned using Cartesian coordinates. We therefore compute a posture-normalized version of each fitted mesh M i in the following way: we start with a mean shape M computed over all M i and use [38] to optimize the localized Laplacian coordinates of M to be as close as possible to M i . This leads to fittings that have the body shape of M i in the common normalized posture of M.…”

Section: Posture Normalizationmentioning

confidence: 99%

“…ours ours+ WSX Shown are eigenvectors of the S-SCAPE space [18] (row 1) and the S-SCAPE spaces trained using our pre-processed data without (row 2) and with posture normalization using WSX [38] (row 3) and NH [21] (row 4).…”

Section: S-scape [18]mentioning

confidence: 99%

“…This representation makes the model versatile and computationally efficient.Prior to statistical analysis, the human scans have to be processed and aligned to establish correspondence. We contribute by evaluating different variants of the state-of-the-art techniques for non-rigid template fitting and posture normalization to process the raw data [1,16,38,21]. Our findings are not entirely new methods, but best practices and specific solutions for automatic preprocessing of large scan databases for learning the S-SCAPE model in the best way (Section 3).…”

mentioning

confidence: 99%

See 1 more Smart Citation

Building statistical shape spaces for 3D human modeling

Pishchulin

Wuhrer

Helten

et al. 2017

Pattern Recognition

Self Cite

180

126

View full text Add to dashboard Cite

Statistical models of 3D human shape and pose learned from scan databases have developed into valuable tools to solve a variety of vision and graphics problems.Unfortunately, most publicly available models are of limited expressiveness as they were learned on very small databases that hardly reflect the true variety in human body shapes. In this paper, we contribute by rebuilding a widely used statistical body representation from the largest commercially available scan database, and making the resulting model available to the community (visit http: // humanshape. mpi-inf. mpg. de ). As preprocessing several thousand scans for learning the model is a challenge in itself, we contribute by developing robust best practice solutions for scan alignment that quantitatively lead to the best learned models. We make implementations of these preprocessing steps also publicly available. We extensively evaluate the improved accuracy and generality of our new model, and show its improved performance for human body reconstruction from sparse input data.2). Our model is based on a simplified and efficient variant of the SCAPE model [3] (henceforth termed S-SCAPE space) that was described by Jain et al. [18] and used for different applications in computer vision and graphics [18,24,23,17,20], but was never learned from such a complete dataset. This compact shape space learns a probability distribution from a dataset of 3D human laser scans. It models variations due to changes in identity using a principal component analysis (PCA) space, and variations due to pose using a skeleton-based surface skinning approach. This representation makes the model versatile and computationally efficient.Prior to statistical analysis, the human scans have to be processed and aligned to establish correspondence. We contribute by evaluating different variants of the state-of-the-art techniques for non-rigid template fitting and posture normalization to process the raw data [1,16,38,21]. Our findings are not entirely new methods, but best practices and specific solutions for automatic preprocessing of large scan databases for learning the S-SCAPE model in the best way (Section 3). First, shape and posture fitting of an initial shape model to a raw scan prior to non-rigid deformation considerably improves the results.

show abstract

“…As a result, in order to specify a new pose given a specific shape, linear regression is required to learn how a model deforms given a set of joint angles. Wuhrer et al [16] proposed an alternative rigid-invariant surface encoding by expressing the Laplacian coordinates of the surface with respect to local frames of reference. The constructed PCA space jointly models shape and non-rigid pose deformations.…”

Section: Related Workmentioning

confidence: 99%

A Layered Model of Human Body and Garment Deformation

Neophytou

Hilton

2014

2014 2nd International Conference on 3D Vision

View full text Add to dashboard Cite

In this paper we present a framework for learning a three layered model of human shape, pose and garment deformation. 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. The shape and pose deformation layers of the model are trained on a rich dataset of full body 3D scans of human subjects in a variety of poses. The garment deformation layer is trained on animated mesh sequences of dressed actors and relies on a novel technique for human shape and posture estimation under clothing.The key contribution of this paper is that we consider garment deformations as the residual transformations between a naked mesh and the dressed mesh of the same subject.

show abstract

Posture-invariant statistical shape analysis using Laplace operator

Cited by 22 publications

References 23 publications

Map-based exploration of intrinsic shape differences and variability

Map-based exploration of intrinsic shape differences and variability

Building statistical shape spaces for 3D human modeling

A Layered Model of Human Body and Garment Deformation

Contact Info

Product

Resources

About