Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Videos

GARAT: Generative Adversarial Learning for Robust and Accurate Tracking.

Bowen Yao¹, Jing Li¹, Shan Xue²

¹School of Computer Science, Wuhan University, Wuhuan 430072, China.

Neural Networks : the Official Journal of the International Neural Network Society

|February 12, 2022

Summary

This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

Photon-counting detector CT in cardiovascular imaging: technical advances, clinical applications, and future perspectives.

The international journal of cardiovascular imaging·2026

Same author

AdaEndoGS: An Adaptive Enlightening Model for Endoscopy Based on 3D Gaussian Splatting.

IEEE transactions on visualization and computer graphics·2026

Same author

Redundant Glycerol-3-Phosphate Acyltransferases Regulate Thermo-Sensitive Genic Male Sterility in Rice.

Plant biotechnology journal·2026

Same author

Abnormal meiocytes and tapetum 1 (OsAMT1) is essential for proper meiosis and tapetal development in rice.

The Plant journal : for cell and molecular biology·2026

Same author

Highly reconfigurable neuronlike conductive networks through nanophase structure engineering.

Nature communications·2025

Same author

Oxygen-Assisted Direct Synthesis of Twisted Bilayer MoS<sub>2</sub> with Tunable Exciton Lifetime.

ACS nano·2025

Same journal

Anchor-based disentanglement framework for incremental multi-view clustering.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Complex-valued amplitude-phase interference modeling for adversarially robust classification.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

TraNce: Type-aware hypergraph neural network with biological mediators for drug repositioning.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Decentralized ADMM for factorization-based Low-rank matrix estimation.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Memristive neuromorphic circuit design inspired by the neural mechanisms of conditioned fear.

Neural networks : the official journal of the International Neural Network Society·2026

Same journal

Q-learning based asynchronous Boolean control networks stabilization with data loss.

Neural networks : the official journal of the International Neural Network Society·2026

See all related articles

This study introduces a new generative adversarial learning method for robust object tracking, even with significant appearance and environmental changes. By simulating and eliminating distractors, the Siamese network achieves more accurate tracking performance.

Area of Science:

Computer Vision
Artificial Intelligence

Background:

Siamese networks are popular for object tracking but struggle with drastic target appearance and environmental changes.
Existing methods face challenges in scenarios with illumination variations and background clutter.

Purpose of the Study:

To propose a novel generative adversarial learning method for robust object tracking.
To address difficulties in tracking targets undergoing drastic changes in appearance, illumination, and background.

Main Methods:

Incorporated a distractor generator into a traditional Siamese network to simulate and eliminate tracking distractors.
Utilized generalized intersection over union (GIoU) as a training loss for improved bounding box regression accuracy.

Main Results:

Keywords:

Generalized intersection over union Generative adversarial learning Object tracking Siamese network

Related Experiment Videos

Achieved more robust tracking performance by mitigating distractors from the search image.
Demonstrated favorable and state-of-the-art results on five challenging benchmarks compared to other trackers.

Conclusions:

The proposed generative adversarial learning method enhances object tracking robustness against significant environmental and appearance variations.
The integration of GIoU loss contributes to more accurate bounding box regression and overall tracking precision.