Evolution of Modern Deep Learning Methods of Object Recognition
Pritom Hazarika, M Kumari · Research in Computing Science · 2019
Vision is a very complex work of our brain.90% of the information entered to our brain is related to vision.In computer vision engineers and scientists are trying to give the computers the ability of vision.Object recognition is one of the most exciting fields of computer vision and AI.There is no rarity of problems and challenges in object recognition, from image classification to keypoint detection.But like many other problems in the world, there is still no obvious and "Best" way to resolve these problems.But the recent advancement in GPU technology propelled the success and accuracy of deep learning algorithms.In this paper, we will study about the evolution of deep learning methods in object recognition.We will also go through some of the classical machine learning methods for object recognition and talk about multimodal approaches to solving these issues.