International Journal of Computer Vision
International Journal of Computer Vision - Special issue on statistical and computational theories of vision: Part II
Statistics and Computing
Hyperdynamics Importance Sampling
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Building Roadmaps of Local Minima of Visual Models
ECCV '02 Proceedings of the 7th European Conference on Computer Vision-Part I
Singularity Analysis for Articulated Object Tracking
CVPR '98 Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
Global and local deformations of solid primitives
SIGGRAPH '84 Proceedings of the 11th annual conference on Computer graphics and interactive techniques
Tracking through Singularities and Discontinuities by Random Sampling
ICCV '99 Proceedings of the International Conference on Computer Vision-Volume 2 - Volume 2
Consistency and Coupling in Human Model Likelihoods
FGR '02 Proceedings of the Fifth IEEE International Conference on Automatic Face and Gesture Recognition
Wormholes in Shape Space: Tracking through Discontinuous Changes in Shape
ICCV '98 Proceedings of the Sixth International Conference on Computer Vision
Maintaining Multi-Modality through Mixture Tracking
ICCV '03 Proceedings of the Ninth IEEE International Conference on Computer Vision - Volume 2
Generative modeling for continuous non-linearly embedded visual inference
ICML '04 Proceedings of the twenty-first international conference on Machine learning
Variational mixture smoothing for non-linear dynamical systems
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Kinematic jump processes for monocular 3D human tracking
CVPR'03 Proceedings of the 2003 IEEE computer society conference on Computer vision and pattern recognition
Approximate Bayesian Multibody Tracking
IEEE Transactions on Pattern Analysis and Machine Intelligence
Inverse Kinematics Using Sequential Monte Carlo Methods
AMDO '08 Proceedings of the 5th international conference on Articulated Motion and Deformable Objects
Hi-index | 0.00 |
Sequential random sampling ('Markov Chain Monte-Carlo') is a popular strategy for many vision problems involving multi-modal distributions over high-dimensional parameter spaces. It applies both to importance sampling (where one wants to sample points according to their 'importance' for some calculation, but otherwise fairly) and to global-optimization (where one wants to find good minima, or at least good starting points for local minimization, regardless of fairness). Unfortunately, most sequential samplers are very prone to becoming trapped for long periods in unrepresentative local minima, which leads to biased or highly variable estimates. We present a general strategy for reducing MCMC trapping that generalizes Voter's 'hyperdynamic sampling' from computational chemistry. The local gradient and curvature of the input distribution are used to construct an adaptive importance sampler that focuses samples on negative curvature regions that are likely to contain low cost 'transition states' (codimension-1 saddle points representing 'mountain passes' connecting adjacent cost basins). This substantially accelerates inter-basin transition rates while still preserving correct relative transition probabilities. Experimental tests on the difficult problem of 3D articulated human pose estimation from monocular images show significantly enhanced minimum exploration.