Top Banner
Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University
11

Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Apr 01, 2015

Download

Documents

Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Probabilistic Tracking and Recognition

of Non-rigid Hand Motion

Huang Fei, Ian ReidDepartment of Engineering Science

Oxford University

Page 2: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

The Problem

• Simultaneous Tracking and Recognition• Articulation and Self-Occlusion• Cluttered Background Scene and Occlusion

Two Successive Frames From A Video Sequence Two Successive Frames From A Video Sequence

Page 3: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Previous Research

• Kinematic Model v.s. Appearance Model• Toyama & Blake “Metric Mixture Tracker” Merits: -Exemplar v.s. Model -Spatial-Temporal Filtering Disadvantages: -Contour (Edges) v.s. Region (Silhouettes) -Joint Observation Density of Two Independent Processes

Page 4: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Method

• System Diagram of Joint Bayes Filter

Page 5: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

• The Interaction Between Two Components in

Joint Bayes Filter

Page 6: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Discrete Appearance Tracker

• Non-Rigid Appearances v.s. Speech Signal• Assumption: -Representative Hand Appearances -Non-Rigid Motion Observe Markov Dependence• The Aim of Learning: -Exemplar as Shape Tracker Representation -Articulated Human Motion Dynamics

Page 7: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Visualizing Non-Rigid Hand Motion• Local Linear Embedding Algorithm (S.Roweis & L.Saul 2000)

Page 8: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Robust Region Tracker

• Use Probabilistic Colour Histogram Tracker (P.Prez

et.al. ECCV 2002) As Global Region Estimator

Page 9: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

• Tracking Global Region and Articulated Motion

Experiments

Frame 1Frame 1 Frame 2Frame 2 Frame 3Frame 3 Frame 4Frame 4 Frame 5Frame 5

Frame 6Frame 6 Frame 7Frame 7 Frame 8Frame 8 Frame 9Frame 9 Frame 10Frame 10

Frame 11Frame 11 Frame 12Frame 12 Frame 13Frame 13 Frame 14Frame 14 Frame 15Frame 15

Page 10: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

• Coping with Occlusion Clutter

Frame 1Frame 1 Frame 2Frame 2 Frame 3Frame 3 Frame 4Frame 4 Frame 5Frame 5

Frame 6Frame 6 Frame 7Frame 7 Frame 8Frame 8 Frame 9Frame 9 Frame 10Frame 10

Frame 11Frame 11 Frame 12Frame 12 Frame 13Frame 13 Frame 14Frame 14 Frame 15Frame 15

Page 11: Probabilistic Tracking and Recognition of Non-rigid Hand Motion Huang Fei, Ian Reid Department of Engineering Science Oxford University.

Conclusion

• Two Independent Dynamic Processes • Two Bayesian Tracker =>Joint Bayes Filter• Robust Global Region Estimator• Robust State-Based Inference