# Adaptive Decontamination of the Training Set: A Unified Formulation for Discriminative Visual Tracking

arXiv: https://arxiv.org/abs/1609.06118 · HTML: https://mlsanity.com/paper/1609.06118/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Sep 2016.

## Built on by (most cited first, 12 of 12)

- [Learning Target Candidate Association to Keep Track of What Not to Track](https://mlsanity.com/paper/2103.16556/) (2021)
- [Learning Occlusion-Robust Vision Transformers for Real-Time UAV Tracking](https://mlsanity.com/paper/2504.09228/) (2025)
- [Dynamic Attention guided Multi-Trajectory Analysis for Single Object Tracking](https://mlsanity.com/paper/2103.16086/) (2021)
- [Siamese Transformer Pyramid Networks for Real-Time UAV Tracking](https://mlsanity.com/paper/2110.08822/) (2021)
- [Learning Residue-Aware Correlation Filters and Refining Scale Estimates with the GrabCut for Real-Time UAV Tracking](https://mlsanity.com/paper/2104.03114/) (2021)
- [Towards Discriminative Representations with Contrastive Instances for Real-Time UAV Tracking](https://mlsanity.com/paper/2308.11450/) (2023)
- [Rank-Based Filter Pruning for Real-Time UAV Tracking](https://mlsanity.com/paper/2207.01768/) (2022)
- [UHP-SOT: An Unsupervised High-Performance Single Object Tracker](https://mlsanity.com/paper/2110.01812/) (2021)
- [Learning Disentangled Representation with Mutual Information Maximization for Real-Time UAV Tracking](https://mlsanity.com/paper/2308.10262/) (2023)
- [DeepMix: Online Auto Data Augmentation for Robust Visual Object Tracking](https://mlsanity.com/paper/2104.11585/) (2021)
- [Deep Convolutional Correlation Iterative Particle Filter for Visual Tracking](https://mlsanity.com/paper/2107.02984/) (2021)
- [Sparse Regularized Correlation Filter for UAV Object Tracking with adaptive Contextual Learning and Keyfilter Selection](https://mlsanity.com/paper/2205.03627/) (2022)

## Abstract (arXiv)

Tracking-by-detection methods have demonstrated competitive performance in recent years. In these approaches, the tracking model heavily relies on the quality of the training set. Due to the limited amount of labeled training data, additional samples need to be extracted and labeled by the tracker itself. This often leads to the inclusion of corrupted training samples, due to occlusions, misalignments and other perturbations. Existing tracking-by-detection methods either ignore this problem, or employ a separate component for managing the training set. We propose a novel generic approach for alleviating the problem of corrupted training samples in tracking-by-detection frameworks. Our approach dynamically manages the training set by estimating the quality of the samples. Contrary to existing approaches, we propose a unified formulation by minimizing a single loss over both the target appearance model and the sample quality weights. The joint formulation enables corrupted samples to be down-weighted while increasing the impact of correct ones. Experiments are performed on three benchmarks: OTB-2015 with 100 videos, VOT-2015 with 60 videos, and Temple-Color with 128 videos. On the OTB-2015, our unified formulation significantly improves the baseline, with a gain of 3.8% in mean overlap precision. Finally, our method achieves state-of-the-art results on all three datasets. Code and supplementary material are available at http://www.cvl.isy.liu.se/research/objrec/visualtracking/decontrack/index.html .