Author : Harpreet Singh 1
Date of Publication :22nd February 2018
Abstract: Multiclass object recognition and classification from the video stream is active research topic in computer vision due to its wide range of application in many emerging areas such as surveillance, medical, safety, vehicle detection. Object recognition and classification task are far more challenging because of image and video data are of heavy and highly variable in nature and harsh nature of real-world recognition and classification scenarios. The processing of image and video data is required to be in real-time. The objective of this paper presents a comprehensive qualitative and quantitative comparative study of several state-of-the-art object recognition and classification methods. We have also examined merits, demerits efficiency of pioneering machine learning methods being used for object recognition and classification.
Reference :