Parametric reshaping of human bodies in images

Zhou, Shizhe; Fu, Hengzhi; Liu, Ligang; Cohen–Or, Daniel; Han, Xiaodong

doi:10.1145/1833349.1778863

Cited by 54 publications

(38 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Here we follow a similar idea, however in our case we use a generative 3D human shape model and motion capture data to generate possible deformations of 2D data making our deformation model more realistic and versatile. [24,32] are probably closest to our work. In our own prior work [24] we require an expensive data acquisition step limiting the number of subjects in the experiments to a handful of people.…”

Section: Introductionsupporting

confidence: 66%

Articulated people detection and pose estimation: Reshaping the future

Pishchulin

Jain

Andriluka

et al. 2012

2012 IEEE Conference on Computer Vision and Pattern Recognition

211

157

View full text Add to dashboard Cite

State-of-the-art methods for human detection and pose estimation require many training samples for best performance. While large, manually collected datasets exist, the captured variations w.r.t. appearance, shape and pose are often uncontrolled thus limiting the overall performance. In order to overcome this limitation we propose a new technique to extend an existing training set that allows to explicitly control pose and shape variations. For this we build on recent advances in computer graphics to generate samples with realistic appearance and background while modifying body shape and pose. We validate the effectiveness of our approach on the task of articulated human detection and articulated pose estimation. We report close to state of the art results on the popular Image Parsing [25] human pose estimation benchmark and demonstrate superior performance for articulated human detection. In addition we define a new challenge of combined articulated human detection and pose estimation in real-world scenes.

show abstract

Section: Introductionsupporting

confidence: 66%

Articulated people detection and pose estimation: Reshaping the future

Pishchulin

Jain

Andriluka

et al. 2012

2012 IEEE Conference on Computer Vision and Pattern Recognition

211

157

View full text Add to dashboard Cite

show abstract

“…More specifically, the SCAPE model allows us to generate an individual body shape by giving θ, and with a pose by giving β. SCAPE model should be learned from a database of human shape with different individuals and different poses. We follow Zhou et al [3] to learn it from a public database [16]. In our case, θ A R 10 and β A R 20 , which cover well the human subspace spanned by the training data.…”

Section: Statistical Model Of Human Bodymentioning

confidence: 99%

“…In our case, θ A R 10 and β A R 20 , which cover well the human subspace spanned by the training data. We refer readers to [15,3] for more details about the definition and training of the SCAPE model.…”

Section: Statistical Model Of Human Bodymentioning

confidence: 99%

See 1 more Smart Citation

Estimation of human body shape and cloth field in front of a kinect

et al. 2015

View full text Add to dashboard Cite

“…Different from those methods estimating a human model from 3D data, Balan et al [4,5] use a stochastic search method to estimate 3D body model from 2D images of multiview. Later on, researchers applied SCAPE model with semantic information to estimate the human body shape and pose from a single image for virtual try on [6] and image body reshaping [7] usage. In our system, SCAPE is used to estimate both pose and shape from 3D data for each frame.…”

Section: Related Workmentioning

confidence: 99%