A survey of advances in vision-based human motion capture and analysis

Moeslund, Thomas B.; Hilton, Adrian; Krüger, Volker

doi:10.1016/j.cviu.2006.08.002

Cited by 2,150 publications

(1,182 citation statements)

References 222 publications

Supporting

Mentioning

1,111

Contrasting

Unclassified

Order By: Relevance

“…Generative models are very popular in the machine learning community, with many variations in existence [e.g. Roweis and Ghahramani 1999, Bishop 2006, Buxton 2003]. …”

Section: Generativementioning

confidence: 99%

Prediction of intent in robotics and multi-agent systems

Demiris

2007

Cogn Process

107

View full text Add to dashboard Cite

Moving beyond the stimulus contained in observable agent behaviour, i.e. understanding the underlying intent of the observed agent is of immense interest in a variety of domains that involve collaborative and competitive scenarios, for example assistive robotics, computer games, robot-human interaction, decision support and intelligent tutoring. This review paper examines approaches for performing action recognition and prediction of intent from a multidisciplinary perspective, in both single robot and multi-agent scenarios, and analyses the underlying challenges, focusing mainly on generative approaches.

show abstract

“…Generative models are very popular in the machine learning community, with many variations in existence [e.g. Roweis and Ghahramani 1999, Bishop 2006, Buxton 2003]. …”

Section: Generativementioning

confidence: 99%

Prediction of intent in robotics and multi-agent systems

Demiris

2007

Cogn Process

107

View full text Add to dashboard Cite

show abstract

“…In short, a significant amount of research has proposed different methods for extracting skeleton from the mesh [4], [7], [8]. However, most of recent research has focused only to draw the line-skeleton in a given model.…”

Section: Motion Capturing and Retargettingmentioning

confidence: 99%

“…The curve-skeleton is useful in shape analysis [6], 3D character animation [7], morphing and shape registration [8], object matching and shape retrieval [9] and surface reconstruction [10].…”

Section: Related Workmentioning

confidence: 99%

Automatic Conversion of Human Mesh into Skeleton Animation by Using Kinect Motion

Razzaq¹,

Wu²,

Zhou³

et al. 2015

IJCTE

View full text Add to dashboard Cite

Abstract-Skeleton-base-animation methods have been commonly used in the character animations. The process of making skeleton for character animation is a long-winded task requiring manual tweaking. This paper presents a novel method to create an automatic animated skeleton from 3D human geometric model through mesh contraction. An automatically generated skeleton is animated by using Kinect captured human motion. The method first, extract a 1D curve-skeleton from the mesh through mesh contraction with constraints. Secondly, the hierarchical joint-based skeleton (armature) has been generated, using the extracted 1D curve-skeleton of the input mesh automatically. Third, the real-time human motion is captured by using the Kinect device. The Kinect captured motion also converts into a standard skeleton motion BVH (Biovision hierarchical) format. Finally, the Kinect motion is retargeted to animate the resulting skeleton of the mesh through joint mapping. The main objective of the proposed approach is to minimize labor-intensive process of skeleton adjustment for character animation. The results of a mesh generated skeleton and plausible skeleton animation to demonstrate the efficiency of the proposed work. The mesh generated skeleton and Kinect motion skeleton both can be directly useful for mesh skinning, mesh rigging, and motion retargeting to create satisfactory character animation.

show abstract

“…Many existing approaches [8] rely on previously extracted silhouettes, for instance by background subtraction, and establish correspondences between contour points and points on the model surface. This involves a contour extraction and a matching of the projected surface with the contour.…”

Section: Region Based Pose Trackingmentioning

confidence: 99%

“…Since there are no artificial markers as in marker based tracking systems, one usually assumes a 3D model of the person to be given. A popular representation is a surface model with known joint positions [8]. Multi-view image streams, i.e., images from synchronized and calibrated cameras, are very common to reduce ambiguities.…”

Section: Introductionmentioning

confidence: 99%

Staying Well Grounded in Markerless Motion Capture

Rosenhahn

Schmaltz

Brox³

et al. 2008

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In order to overcome typical problems in markerless motion capture from video, such as ambiguities, noise, and occlusions, many techniques reduce the high dimensional search space by integration of prior information about the movement pattern or scene. In this work, we present an approach in which geometric prior information about the floor location is integrated in the pose tracking process. We penalize poses in which body parts intersect the ground plane by employing soft constraints in the pose estimation framework. Experiments with rigid objects and the HumanEVA-II benchmark show that tracking is remarkably stabilized.

show abstract

A survey of advances in vision-based human motion capture and analysis

Cited by 2,150 publications

References 222 publications

Prediction of intent in robotics and multi-agent systems

Prediction of intent in robotics and multi-agent systems

Automatic Conversion of Human Mesh into Skeleton Animation by Using Kinect Motion

Staying Well Grounded in Markerless Motion Capture

Contact Info

Product

Resources

About