Online adaptive hidden Markov model for multi-tracker fusion

@article{Vojr2016OnlineAH,
  title={Online adaptive hidden Markov model for multi-tracker fusion},
  author={Tom{\'a}s Voj{\'i}r and Jiri Matas and Jana Noskova},
  journal={ArXiv},
  year={2016},
  volume={abs/1504.06103}
}
A Distilled Model for Tracking and Tracker Fusion
TLDR
This paper proposes a novel tracking methodology that takes advantage of other visual trackers, offline and online, and shows that the proposed algorithms compete with state-of-the-art trackers while running in real-time.
Tracking-by-Trackers with a Distilled and Reinforced Model
TLDR
This paper proposes a novel tracking methodology that takes advantage of other visual trackers, offline and online, and extensive validation shows that the proposed algorithms compete with real-time state-of-the-art trackers.
Exploring bounding box context for multi-object tracker fusion
TLDR
This paper analyzes three current state-of-the-art MOT trackers and shows that there is still room for improvement, and performs an extensive evaluation over ten common evaluation sequences to consistently show improved performances.
Re3 : Real-Time Recurrent Regression Networks for Object Tracking
TLDR
Re is presented, a real-time deep object tracker capable of incorporating long-term temporal information into its model, using a recurrent neural network to represent the appearance and motion of the object.
Real-Time Ensemble-Based Tracker with Kalman Filter
This work presents an ensemble-based visual object tracker called KFebT. This method can fuse using a Kalman Filter the result of several out-of-the box trackers or specialist methods that solve
Single-target tracking of arbitrary objects using multi-layered features and contextual information
TLDR
A novel RGB single-target tracker which models the target with multi-layered features and contextual information exploiting global dynamic constraints from the context is proposed and demonstrated that the proposed method outperforms other state-of-the-art single target trackers on highly cluttered scenes.
Real-time least-squares ensemble visual tracking
TLDR
A novel ensemble tracking system by formulating the tracking task in terms of a linear regression which is a least-squares problem and a novel sampling strategy is proposed to train accurate and diverse weak classifiers.
CoCoLoT: Combining Complementary Trackers in Long-Term Visual Tracking
TLDR
CoLoT perceives whether the trackers are following the target object through an online learned deep verification model, and accordingly activates a decision policy which selects the best performing tracker as well as it corrects the performance of the failing one.
On-line fusion of trackers for single-object tracking
...
...

References

SHOWING 1-10 OF 39 REFERENCES
Visual Tracking via Adaptive Tracker Selection with Multiple Features
TLDR
A robust visual tracking method to track an object in dynamic conditions that include motion blur, illumination changes, pose variations, and occlusions is proposed, which retains the robustness and adaptability of the TLF and multiple trackers.
PROST: Parallel robust online simple tracking
TLDR
This work shows that augmenting an on-line learning method with complementary tracking approaches can lead to more stable results, and uses a simple template model as a non-adaptive and thus stable component, a novel optical-flow-based mean-shift tracker as highly adaptive element and anon-line random forest as moderately adaptive appearance-based learner.
MEEM: Robust Tracking via Multiple Experts Using Entropy Minimization
TLDR
It is shown that the proposed multi-expert restoration scheme significantly improves the robustness of the base tracker, especially in scenarios with frequent occlusions and repetitive appearance variations.
Tracking-Learning-Detection
  • Zdenek Kalal
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2012
TLDR
A novel tracking framework (TLD) that explicitly decomposes the long-term tracking task into tracking, learning, and detection, and develops a novel learning method (P-N learning) which estimates the errors by a pair of “experts”: P-expert estimates missed detections, and N-ex Expert estimates false alarms.
Ensemble-Based Tracking: Aggregating Crowdsourced Structured Time Series Data
TLDR
This work proposes a factorial hidden Markov model (FHMM) for ensemble-based tracking by learning jointly the unknown trajectory of the target and the reliability of each tracker in the ensemble and devise a conditional particle filter algorithm by exploiting the structure of the joint posterior distribution of the hidden variables.
Object Tracking by Oversampling Local Features
  • F. Pernici, A. Bimbo
  • Computer Science
    IEEE Transactions on Pattern Analysis and Machine Intelligence
  • 2014
TLDR
This paper presents the ALIEN tracking method that exploits oversampling of local invariant representations to build a robust object/context discriminative classifier and shows that the learning rule has asymptotic stability under mild conditions and confirms the drift-free capability of the method in long-term tracking.
An Enhanced Adaptive Coupled-Layer LGTracker++
TLDR
This paper identifies four important cases when the original tracker often fails: significant scale changes, environment clutter, and failures due to occlusion and rapid disordered movement, and suggests four new enhancements to solve these problems.
A Novel Performance Evaluation Methodology for Single-Target Trackers
TLDR
The requirements are the basis of a new evaluation methodology that aims at a simple and easily interpretable tracker comparison and a fully-annotated dataset with per-frame annotations with several visual attributes, which is the largest benchmark to date.
Visual Object Tracking by Structure Complexity Coefficients
TLDR
A novel visual object tracking algorithm based on the observation dependent hidden Markov model (OD-HMM) framework and the observation dependency is computed by structure complexity coefficients (SCC) which is defined to predict the target appearance change.
Robust Visual Tracking Using an Adaptive Coupled-Layer Visual Model
TLDR
A novel coupled-layer visual model that combines the target's global and local appearance by interlacing two layers is proposed that outperforms the related trackers by having a smaller failure rate as well as better accuracy.
...
...