Visual tracking based on transfer learning of deep salience information

Journal Title: Opto-Electronic Advances - Year 2020, Vol 3, Issue 9

Abstract

In this paper, we propose a new visual tracking method in light of salience information and deep learning. Salience detection is used to exploit features with salient information of the image. Complicated representations of image features can be gained by the function of every layer in convolution neural network (CNN). The characteristic of biology vision in attention-based salience is similar to the neuroscience features of convolution neural network. This motivates us to improve the representation ability of CNN with functions of salience detection. We adopt the fully-convolution networks (FCNs) to perform salience detection. We take parts of the network structure to perform salience extraction, which promotes the classification ability of the model. The network we propose shows great performance in tracking with the salient information. Compared with other excellent algorithms, our algorithm can track the target better in the open tracking datasets. We realize the 0.5592 accuracy on visual object tracking 2015 (VOT15) dataset. For unmanned aerial vehicle 123 (UAV123) dataset, the precision and success rate of our tracker is 0.710 and 0.429.

Authors and Affiliations

Haorui Zuo*, Zhiyong Xu*, Jianlin Zhang, Ge Jia

Keywords

Related Articles

Large Rabi splitting obtained in Ag-WS2 strong-coupling heterostructure with optical microcavity at room temperature

Manipulation of light-matter interaction is critical in modern physics, especially in the strong coupling regime, where the generated half-light, half-matter bosonic quasiparticles as polaritons are important for fundame...

Acoustic wave detection of laser shock peening

In order to overcome the existing disadvantages of offline laser shock peening detection methods, an online detection method based on acoustic wave signals energy is provided. During the laser shock peening, an acoustic...

Compact pulsed thulium-doped fiber laser for topographical patterning of hydrogels

We report the generation of high energy 2 μm picosecond pulses from a thulium-doped fiber master oscillator power amplifier system. The all-fiber configuration was realized by a flexible large-mode area photonic crystal...

Silicon photonics for telecom and data-com applications

In recent decades, silicon photonics has attracted much attention in telecom and data-com areas. Constituted of high refractive-index contrast waveguides on silicon-on-insulator (SOI), a variety of integrated photonic pa...

Movable electrowetting optofluidic lens for optical axial scanning in microscopy

Optical axial scanning is essential process to obtain 3D information of biological specimens. To realize optical axial scanning without moving, the tunable lens is a solution. However, the conventional tunable lenses usu...

Download PDF file
  • EP ID EP693127
  • DOI 10.29026/oea.2020.190018
  • Views 76
  • Downloads 0

How To Cite

Haorui Zuo*, Zhiyong Xu*, Jianlin Zhang, Ge Jia (2020). Visual tracking based on transfer learning of deep salience information. Opto-Electronic Advances, 3(9), -. https://europub.co.uk/articles/-A-693127