Manuscript Topics

Deep learning approaches have been demonstrated to outshine prior state-of-the-art machine-learning algorithms in various disciplines in recent times, in which the computer vision domain become more prevalent. The computer vision domain is emerged more due to deep learning technologies in a range of application fields, including object identification, semantic segmentation, classification, tracking, and localization. Similarly, Deep reinforcement learning (DRL), on the other hand, has been gradually adopted in a variety of applications such as finance, recommendation systems, video games, and many more. More recent research theories have stretched the routes of DRL to the world of computer vision. Various kinds of DRL frameworks have been designed to handle the problem of landmark detection, object detection, tracking, image registration, and segmentation as well as video data analytics. Furthermore, robot navigation, action recognition, and object searching using a mix of visual and DRL systems will result in more advanced and comprehensive automated systems.

This special issue is concerned with the application of deep reinforcement learning frameworks in the computer vision domain. More specifically, in computer vision DRL can be leveraged for medical imaging problems, for instance, lesion or abnormalities localization, and registration. In addition, they can be also utilized for hyperparameters optimization including the suitable configuration of deep learning architecture for several computer vision problems. Data scarcity, which can be addressed with data augmentation in computer vision, can also be addressed with robust learning of model-free DRL frameworks. Problems with selecting distinct subsets of features taken from images using reinforcement learning are also conceivable, which can previously be done using other feature selection approaches such as LDA and PCA. Such RL-based agents or expert systems can also be expanded to other computer vision applications. We invite researchers and scientists working on reinforcement learning to come up with innovative ideas for integrating these frameworks in the domain of computer vision to solve different more effectively.

Topics of interest include, but are not limited to:
• Application of deep reinforcement learning in landmark and object detection
• Feature selection methods using reinforcement learning.
• Visual question generation using deep reinforcement learning.
• RL-agents for Image captioning
• Automating industrial process using RL agents with visual inputs
• Handling class imbalance and image enhancement problems of computer vision and deep learning using reinforcement learning
• Image hashing and restoration using reinforcement learning.
• Vision-based navigation systems using reinforcement learning.
• Future of computer vision and reinforcement learning
• Reinforcement learning models for content-based image retrievals systems
• Application of DRL in segmentation problems of medical imaging
• Hyperspectral image analysis using reinforcement learning
• Reinforcement learning in remote sensing interpretations
• Vision-based dialogue systems using reinforcement learning
• Reinforcement learning in computer-vision assisted robot navigation problems
• IoT-based agriculture management using RL-agents
• Path planning of AUVs using reinforcement learning
• Optimizing deep learning architectures for computer vision problem using reinforcement learning
• Adversarial robustness of deep reinforcement learning models in computer vision

Instructions for authors
https://www.aimspress.com/era/news/solo-detail/instructionsforauthors
Please submit your manuscript to online submission system
https://aimspress.jams.pub/