Robust Object Tracking with Online Multiple Instance Learning Advisor: Sheng-Jyh Wang Student: Pei Chu Boris Babenko, Ming-Hsuan Yang, Serge Belongie. Robust Object Tracking with Online Multiple Instance Learning. IEEE Trans. on PAMI , 2011.
Dec 13, 2015
Robust Object Tracking with Online Multiple Instance Learning
Advisor: Sheng-Jyh WangStudent: Pei Chu
Boris Babenko, Ming-Hsuan Yang, Serge Belongie. Robust Object Tracking with Online Multiple Instance Learning. IEEE Trans. on PAMI , 2011.
Outline
• Introduction• Tracking by Detection( Related Work)• Multiple Instance Learning (MIL)• Online MILboost• Experiments• Conclusion
Introduction: Tracking
• Problem: track arbitrary object in video given location in first frame
• Typical Tracking System:• Appearance Model
• Color , subspaces, feature,etc• Optimization/Search
• Greedy local search, etc
[Ross et al. ‘07]
Tracking by Detection
• Recent tracking work• Focus on appearance model• Borrow techniques from object detection
• Slide a discriminative classifier around image
[Collins et al. ‘05, Grabner et al. ’06, Ross et al. ‘08]
Tracking by Detection: Online AdaBoost
• Grab one positive patch, and some negative patch, and train/update the model.
negative
positive
Classifier Online classifier (i.e. Online AdaBoost)
Tracking by Detection
• Find max response
negative
positive old location
new location
XX
ClassifierClassifier
Tracking by Detection
• Repeat…
negative
positive
negative
positive
ClassifierClassifier
Problems
• What if classifier is a bit off?• Tracker starts to drift
• How to choose training examples?
Multiple Instance Learning (MIL)
• Instead of instance, get bag of instances• Bag is positive if one or more of it’s members is positive
[Keeler ‘90, Dietterich et al. ‘97] [Viola et al. ‘05]
Positive
Negative
Multiple Instance Learning (MIL)
• MIL Training Input
• The bag labels are defined as:
Online MILBoostFrame t Frame t+1
Get data (bags)
Update all M classifiersin pool
Greedily add best K tostrong classifier
Boosting
• Train classifier of the form:
where is a weak classifier• Can make binary predictions using
[Freund et al. ‘97]
Online MILBoost
• At t frame, Update all M candidate classifiers
• Pick best K in a greedy fashion (M>>K)
[Grabner et al. ‘06]
Online MILBoost
• Objective to maximize: Log likelihood of bags:
where:
[Viola et al. ’05, Friedman et al. ‘00]
Noisy-OR Model, The bag probability
The instance probability
Online MILBoost( OMB)
M>K,
M :is total weak classifier candidates
K : is choosing the best K classifiers
Online MILBoost VS Online Adaboost
System Overview: MILtrack
Experiments
• Compare MILTrack to:• OAB1 = Online AdaBoost w/ 1 pos. per frame• OAB5 = Online AdaBoost w/ 45 pos. per frame• SemiBoost = Online Semi-supervised Boosting• FragTrack = Static appearance model
[Grabner ‘06, Adam ‘06, Grabner ’08]
Results
Results
ResultsBestSecond Best
Conclusions
• Proposed Online MILBoost algorithm• Using MIL to train an appearance model results
in more robust tracking