Enseignant chercheur
Publié le
Thématiques de recherche : Currently, my work focuses on developing new methods to improve the efficiency of semantic image segmentation networks. My goal is to reduce the computational complexity of models by dynamically adapting the segmentation process based on the information density of different regions of the image. This approach aims to optimize resource usage while maintaining high segmentation performance. The project seeks to provide a lighter solution for low-resource applications. I am also collaborating with other researchers to refine this method and evaluate its potential across different use cases.
Points forts des activités de recherche : The strengths of my research focus on optimizing semantic image segmentation models by reducing their complexity while maintaining high performance. I employ techniques such as dynamic gating and adapting the segmentation process based on the information density of image regions, which allows for better allocation of computational resources. This approach targets low-resource applications, such as mobile devices and embedded systems, where efficiency is crucial. I use advanced technologies like CNNs, deep learning, and others to enhance system performance while reducing computational footprint.
Page perso : https://mosakkari.github.io/My-Site/
