Shape-from-silhouette of articulated objects and its use for human body kinematics estimation and motion capture

By German K. M. Cheung, Simon Baker, Takeo Kanade. 2003 IEEE Computer Society Conference on Computer Vision and Pattern Recognition, 2003. Proceedings., 2003

420 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/07c1f9b74eb9a419b9ad5af083c74d00499aacbd

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Motion Interchange Patterns for Action Recognition in Unconstrained Videos (2012, background): “Several early attempts relying on high-level information, include explicit models of human bodies in motion [3], silhouettes [4] or 3D volumes [5].”
Action recognition in videos acquired by a moving camera using motion decomposition of Lagrangian particle trajectories (2011, background): “Aside from trajectory features, a variety of feature representations have been developed for action recognition such as appearance features [9], shape-based representation [10]), volumetric features (e.”
Motionlets: Mid-level 3D Parts for Human Motion Recognition (2013, background): “But the global nature makes them sensitive to intra-class variation and deformation.”
A review on vision techniques applied to Human Behaviour Analysis for Ambient-Assisted Living (2012, methodology): “In indirect model use techniques, in turn, methods that use direct reconstruction of both model shape and motion from the visual-hull are revealed [23, 70].”
Learning to Reconstruct People in Clothing From a Single RGB Camera (2019, background): “[17, 18], where they align visual hulls over time to improve shape estimation.”
Learning View-Invariant Sparse Representations for Cross-View Action Recognition (2013, background): “In order to accurately recognize human actions, most existing approaches focus on developing different discriminative features, such as spatio-temporal interest point (STIP) based features [32, 2, 13, 19], shape [16, 3, 21] and optical flow based features [5, 17, 16].”
Vision-based human motion analysis: An overview (2007, methodology): “We describe model appearance in the image domain, and the matching functions, in Section 2.2.”
Human computing and machine understanding of human behavior: a survey (2006, methodology): “Human Motion Tracking based on 3D Human Kinematic Model [11]”
Fast multiple object tracking via a hierarchical particle filter (2005, positive): “Model-based tracking algorithms incorporate a priori information about the objects to develop representations such as skin complexion, body blobs, kinematic skeleton, silhouettes or layer information [38, 36, 5, 33, 32, 6].”
Efficient mean-shift tracking via a new similarity measure (2005, methodology): “Knowledge-based tracking algorithms incorporate a priori information about the tracked objects to obtain representations such as projected shape, skin complexion, body blobs, kinematic skeletons and silhouettes [36, 34, 5, 30, 7 ]. Learning-based approaches apply pattern recognition algorithms to l”
Human Action Recognition in Videos Using Kinematic Features and Multiple Instance Learning (2010, neutral): “The popular shape-based representations include edges [19] and silhouettes of the human body [22].”
Recognizing human actions using multiple features (2008, background): “The popular shape based representations include edges [15] and silhouettes of human body [18].”
SCAPE: shape completion and animation of people (2005, methodology): “Thus, if we could solve the data acquisition problem (e.g., using shape from silhouette [Cheung et al. 2003]), we could use this framework to learn realistic deformation models for creatures other than humans.”
Performance animation from low-dimensional control signals (2005, methodology): “For example, local models could be used to constrain the search space for markerless human motion capture [Cheung et al. 2003] as well as motion planning [Yamane et al.”
Evaluating New Variants of Motion Interchange Patterns (2013, background): “One line of research extracts the high-level information of the human shape in motion by building explicit models of bodies [30], silhouettes [2] or 3D volumes [6].”
Cross-view action recognition via view knowledge transfer (2011, background): “Some leading representations include space-time pattern templates [3, 27], shape features [36, 24, 7, 19], interest point based representations [15, 26, 29, 20], learned geometrical models of the human body parts [12], and motion/optical flow patterns [1, 25, 36].”
Video Based Reconstruction of 3D People Models (2018, background): “Our work is also related to the seminal work of [14, 15] where they align visual hulls over time to improve shape estimation.”