On the use of Deep Reinforcement Learning for Visual Tracking: a Survey
| Autor: | Marco La Cascia, Giorgio Cruciata, Liliana Lo Presti |
|---|---|
| Přispěvatelé: | Cruciata Giorgio, Lo Presti Liliana, La Cascia Marco |
| Jazyk: | angličtina |
| Rok vydání: | 2021 |
| Předmět: |
General Computer Science
Computer science Feature extraction Machine learning computer.software_genre Field (computer science) video-surveillance Minimum bounding box Reinforcement learning General Materials Science Settore ING-INF/05 - Sistemi Di Elaborazione Delle Informazioni deep reinforcement learning Computer vision machine learning video-surveillance deep reinforcement learning visual tracking business.industry General Engineering Tracking system visual tracking Visualization Active appearance model TK1-9971 machine learning Eye tracking Computer vision Artificial intelligence Electrical engineering. Electronics. Nuclear engineering business computer |
| Zdroj: | IEEE Access, Vol 9, Pp 120880-120900 (2021) |
| Popis: | This paper aims at highlighting cutting-edge research results in the field of visual tracking by deep reinforcement learning. Deep reinforcement learning (DRL) is an emerging area combining recent progress in deep and reinforcement learning. It is showing interesting results in the computer vision field and, recently, it has been applied to the visual tracking problem yielding to the rapid development of novel tracking strategies. After providing an introduction to reinforcement learning, this paper compares recent visual tracking approaches based on deep reinforcement learning. Analysis of the state-of-the-art suggests that reinforcement learning allows modeling varying parts of the tracking system including target bounding box regression, appearance model selection, and tracking hyper-parameter optimization. The DRL framework is elegant and intriguing, and most of the DRL-based trackers achieve state-of-the-art results. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|