Sciweavers

83 search results - page 6 / 17
» 3D Occlusion Inference from Silhouette Cues
Sort
View
MVA
2002
149views Computer Vision» more  MVA 2002»
13 years 7 months ago
3-Dimensional Tracking of Multiple Object Motions from Multi-view Images by Mixed-state CONDENSATION Algorithm
In this paper, we propose an efficient method for tracking multiple objects, which deals with occlusion situations in an image. We use Shape from silhouettes method to reconstruct...
Koichiro Deguchi, Koji Hamasaki, Taira Nakajima, T...
IJCV
2007
117views more  IJCV 2007»
13 years 7 months ago
Implicit Meshes for Effective Silhouette Handling
Abstract. Using silhouettes in uncontrolled environments typically requires handling occlusions as well as changing or cluttered backgrounds, which limits the applicability of most...
Slobodan Ilic, Mathieu Salzmann, Pascal Fua
CVPR
2006
IEEE
14 years 1 months ago
Learning Joint Top-Down and Bottom-up Processes for 3D Visual Inference
We present an algorithm for jointly learning a consistent bidirectional generative-recognition model that combines top-down and bottom-up processing for monocular 3d human motion ...
Cristian Sminchisescu, Atul Kanaujia, Dimitris N. ...
ACII
2005
Springer
14 years 1 months ago
Using an Avatar to Develop a System for the Predication of Human Body Pose from Moments
Tracking people using movie sequences is not straightforward because of the human body's articulation and the complexity of a person's movements. In this paper we show ho...
Song Hu, Bernard F. Buxton
CVPR
2010
IEEE
14 years 4 months ago
Using cloud shadows to infer scene structure and camera calibration
We explore the use of clouds as a form of structured lighting to capture the 3D structure of outdoor scenes observed over time from a static camera. We derive two cues that relate ...
Nathan Jacobs, Brian Bies, and Robert Pless