Recent Trends on Real Time Object Detection using Single Shot Multibox Detector

Ritika Dhiman, Dr. Jaswanti

Abstract: This research paper investigates the running of object detection algorithm on low-end devices to detect different kinds of objects in images. Deep convolutional neural networks (CNNs) which are used in SSD have recently proven extremely capable of performing object detection in single-frame images. Single shot multi-box object detectors have been recently shown to achieve state-of-the-art performance on object detection tasks. The implementation can be done using Pytorch object detection library, and COCO (Common Objects and Context) or Pascal VOC (Visual Object Classes) dataset to detect the common objects around a person like cars, dogs, laptops, etc. The main advantage of using SSD is that, unlike other methods it can be used in laptops and other personal devices.

Keywords: Computer Vision, Image Procession, Object Detection, CNNs Convolutional Neural Networks, Low-end devices, COCO Common Objects And Context, Pascal VOC Visual Object Classes, Python, Open CV and ROIs Region Of Interest

How to Cite?: Ritika Dhiman, Dr. Jaswanti, "Recent Trends on Real Time Object Detection using Single Shot Multibox Detector", Volume 8 Issue 8, August 2019, International Journal of Science and Research (IJSR), Pages: 1447-1449, https://www.ijsr.net/getabstract.php?paperid=ART2020526, DOI: https://dx.doi.org/10.21275/ART2020526