Deep learning in video multi-object tracking: A survey

By Gioele Ciaparrone, Francisco Luque Sánchez, Siham Tabik, Luigi Troiano, Roberto Tagliaferri, Francisco Herrera. Neurocomputing, 2019

668 citations recorded by the source corpus.

Abstract The problem of Multiple Object Tracking (MOT) consists in following the trajectory of different objects in a sequence, usually a video. In recent years, with the rise of Deep Learning, the algorithms that provide a solution to this problem have benefited from the representational power of deep models. This paper provides a comprehensive survey on works that employ Deep Learning models to solve the task of MOT on single-camera videos. Four main steps in MOT algorithms are identified, and an in-depth review of how Deep Learning was employed in each one of these stages is presented. A complete experimental comparison of the presented works on the three MOTChallenge datasets is also provided, identifying a number of similarities among the top-performing methods and presenting some possible future research directions.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/36973996f54a0dd1baf9e4e49a10db7b4d3a4fb9

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Matteo Bruni
2. Claudio Baecchi
3. Federico Pernici
4. A. del Bimbo
5. Rahaf Aljundi

Citations in the corpus

• A Survey of Sound Source Localization with Deep Learning Methods (2021, neutral): “More recently, deep approaches to the MOT problem have emerged, an evolution mostly driven by the computer vision community [Ciaparrone et al., 2020].”
• Regular Polytope Networks (2021, background): “Clearly, also, multiobject tracking [71], [72] can benefit from memory-based learning.”