Articulated body motion capture by annealed particle filtering

By Jonathan A. J. Deutscher, Andrew Blake, Ian David Reid. Proceedings IEEE Conference on Computer Vision and Pattern Recognition. CVPR 2000 (Cat. No.PR00662), 2000

1,098 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/6f6613ef17d3d7627adf3108ac4b3be9e0abfb6e

How researchers see this paper

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

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Model-based hand tracking using a hierarchical Bayesian filter (2006, background): “Great strides have been made in the theory and practice of tracking , for example the development of particle filters recognized that a key aspect in tracking was a better representation of the posterior distribution of model parameters [11, 17, 20, 36, 37].”
View independent human body pose estimation from a single perspective image (2004, methodology): “Unlike previous approachesthat have useda restrictivecameramodelor assumeda calibratedcamera,our work dealswith the generalcaseof a perspective uncalibratedcameraand is thus well suitedfor archived video.”
Pedestrian detection and tracking with night vision (2005, methodology): “[17], [18], spatial–temporal silhouette analysis of human parts with a three-dimensional (3-D) model [20], periodicity and self-similarity analysis [9], annealed particle filtering for articulated pedestrian [11], and probabilistic modeling of pose and motion, as well as maximum a posteriori detecti”
Learning Joint Top-Down and Bottom-up Processes for 3D Visual Inference (2006, methodology): “Generative human models and efficient inference algorithms are described in [6] [24] [25] [26] [30].”
Strike a pose: tracking people by finding stylized poses (2005, background): “Annealing the particle filter [5] or performing local searches [21] are ways to attack this difficulty.”
Advances in View-Invariant Human Motion Analysis: A Review (2010, background): “[55] which combines a deterministic annealing approach with stochastic sampling to reduce the number of samples required.”
Learning shape models for monocular human pose estimation from the Microsoft Xbox Kinect (2011, background): “2D rectangles [15], ellipses [22] or conic sections [6].”
MoDeep: A Deep Learning Framework Using Motion Features for Human Pose Estimation (2014, background): “Several other model based articulated tracking systems have been reported over the past two decades, most notably [10–16].”
Tracking loose-limbed people (2004, methodology): “As suggested in [5] the annealing process uses 10 layers and 200 particles.”
Finding and tracking people from the bottom up (2003, background): “Annealing the particle filter is one way to attack this difficulty [6].”
3D hand tracking by rapid stochastic gradient descent using a skinning model (2004, methodology): “Subsequent rows present the results of the BroydenFletcher-Goldfarb-Shanno (BFGS) [18], the Gradient Descent (GD) [18], the Powell [18] and the Annealed Particle Filter (APF) [3] algorithms.”
Iterated Second-Order Label Sensitive Pooling for 3D Human Pose Estimation (2014, background): “Many of the early approaches to 3D human pose estimation employed generative models to search the state space in order to identify configurations of a body model that would best align with image features[12, 32, 29, 34, 15].”
Latent structured models for human pose estimation (2011, methodology): “The 2D to 3D imaging relations and their ambiguities in the articulated case under point-wise joint correspondences are relatively well understood both in the monocular and in the multi-camera setting; several classes of methods exist: generative sampling strategies [8, 5], discrimi native methods b”
Free-Viewpoint Video of Human Actors Using Multiple Handheld Kinects (2013, background): “The last are markerless motion capture approaches that concentrate on capturing the skeleton motion of the human performance, e.g., [27], [28], [29], [30], [31].”
Monocular Pedestrian Detection: Survey and Experiments (2009, background): “Extensions that are especially relevant for pedestrian tracking involve hybrid discrete/continuous state-spaces [26], [50] and efficient sampling strategies [13], [32], [36], [44].”
Capturing natural hand articulation (2001, methodology): “A method of combining edge and silhouette observations was reported recently for human body tracking [2].”
Model-based 3D tracking of an articulated hand (2001, methodology): “An annealed version of particle filtering is developed in [ 10 ], which, although producing very good results, has extremely high computational cost.”