Fourier-Information Duality in the Identity Management Problem thumbnail
Pause
Mute
Subtitles
Playback speed
0.25
0.5
0.75
1
1.25
1.5
1.75
2
Full screen

Fourier-Information Duality in the Identity Management Problem

Published on Oct 03, 20112788 Views

We compare two recently proposed approaches for representing probability distributions over the space of permutations in the context of multi-target tracking. We show that these two representations,

Related categories

Chapter list

Fourier-Information Duality in the Identity Management Problem00:00
Problem in Identity Management [Shin et. al. 2003]00:00
Problem in Identity Management01:00
Reasoning over Permutation Group01:41
Contributions02:32
Fourier Approach03:14
Fourier Approach - High Order Generalizations04:18
Information Approach05:28
Information Approach - High Order Generalizations06:16
Two Forms of Representation06:16
Markov Model for Identity Management06:51
Mixing Model (1)07:46
Mixing Model (2)08:30
Mixing Model (3)09:16
Mixing Model - Fourier and Information Form10:20
Observation Model10:40
Observation Model - Fourier and Information Form11:18
Normalization and Maximization11:54
Comparison of the Two Forms12:43
Both Forms are Low-Dimensional Projections13:10
Pythagorean Theorem14:04
Hybrid Approach (1)14:33
Hybrid Approach (2)15:57
Real Camera Data from Simulation18:48
Running Time19:21
Accuracy20:29
Errors in Distribution21:07
More Experiments21:57
Adaptive. vs Nonadaptive22:27
Conclusions22:44