Performance Analysis of Object Detection Algorithms on YouTube Video Object Dataset

Sharma, Chethan and Singh, Siddharth and Poornalatha, G and Shenoy, Ajitha K B (2021) Performance Analysis of Object Detection Algorithms on YouTube Video Object Dataset. Engineering Letters, 29 (2). ISSN 1816-093X

[img] PDF
12347.pdf - Published Version
Restricted to Registered users only

Download (1MB) | Request a copy

Abstract

Object Recognition is a terminology used to refer to a collection of computer vision tasks that are involved in object identification in digital images and videos. In this paper, different object detection algorithms were implemented on Youtube object dataset. Each object detection algorithm has its own advantages and limitations which depend on the dataset used. It was observed that YOLO and SSD, being state-of-art algorithms, demonstrate better performance than other models on youtube video object dataset. SSD is better at detecting smaller objects. Centernet performs poorly on this dataset

Item Type: Article
Uncontrolled Keywords: Average Precision, CenterNet, DETR, Object Recognition, SSD, YOLO.
Subjects: Engineering > MIT Manipal > Information and Communication Technology
Depositing User: MIT Library
Date Deposited: 20 Aug 2021 06:01
Last Modified: 20 Aug 2021 06:01
URI: http://eprints.manipal.edu/id/eprint/157152

Actions (login required)

View Item View Item