Systematic Analysis of a Context-Aware Deep Learning Architecture for Object Detection
A two stage architecture is proposed aimed at learning contextual relationships, and improving the precision of a CNN-based object detector. A toy image generator is implemented, applying a series of pre-defined contextual relationships, and various experiments are conducted to evaluate the effectiveness of this design.
Bachelor in Computer Engineering. +15 years of experience in various IT industry roles. Master in Machine Learning from Universite Jean Monnet, Saint Etienne, France. Advanced master in AI, KU Leuven, 2019.