Visual product search in mobile commerce

Mobile Visual Search (MVS) is an emerging technology which makes use of the images captured by a mobile device’s built-in camera and perform image search to return users with additional and meaningful information. This includes web pages, similar images or documents, 3D view of the objects and etc....

Full description

Saved in:
Bibliographic Details
Main Author: Lok, Seng Siong
Other Authors: Yap Kim Hui
Format: Final Year Project
Language:English
Published: 2017
Subjects:
Online Access:http://hdl.handle.net/10356/72279
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-72279
record_format dspace
spelling sg-ntu-dr.10356-722792023-07-07T17:22:17Z Visual product search in mobile commerce Lok, Seng Siong Yap Kim Hui School of Electrical and Electronic Engineering DRNTU::Engineering::Electrical and electronic engineering Mobile Visual Search (MVS) is an emerging technology which makes use of the images captured by a mobile device’s built-in camera and perform image search to return users with additional and meaningful information. This includes web pages, similar images or documents, 3D view of the objects and etc. On top of that, with the integration of Augmented Reality (AR), visual searching and matching can now be done within seconds from the live camera view of the visual search application rather than taking photos and save as images before any matching to be done at the server end. However, existing technologies are still having limitations which stop MVS from performing as perfect as it aims to be. The physical appearance of an object, lighting conditions, angles of scanning are examples of the objective factors which has limit the performance of MVS. Hence, this led to the necessary of having an alternative way for users to perform a non-text searching when it is required. Subsequently, Quick Response (QR) code and 1D Barcode are chosen to fulfill the objective. In other words, the goal of this project is to integrate both QR code and 1D barcode scanning function into the existing application as well as update the search engine with new data in order for it to recognise barcode or QR code printed on a console game cover and redirect users to the Uniform Resource Locator (URL) of the game trailer. This will certainly allow users to have a better idea of the gameplay before any purchase and the application will also be able to switch to different scanning mode automatically upon user’s selection if the scanning inputs are detected to be invalid. At the end of this project, the search engine was updated with new data and the series of testing were conducted to ensure the integration of 1D barcode and QR code with the visual search are working seamlessly under one application. As a result, the application is now able to provide extra details of the console game after perform scanning on its cover photo using different types of scanning mode. Bachelor of Engineering 2017-05-31T07:31:10Z 2017-05-31T07:31:10Z 2017 Final Year Project (FYP) http://hdl.handle.net/10356/72279 en Nanyang Technological University 54 p. application/pdf
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic DRNTU::Engineering::Electrical and electronic engineering
spellingShingle DRNTU::Engineering::Electrical and electronic engineering
Lok, Seng Siong
Visual product search in mobile commerce
description Mobile Visual Search (MVS) is an emerging technology which makes use of the images captured by a mobile device’s built-in camera and perform image search to return users with additional and meaningful information. This includes web pages, similar images or documents, 3D view of the objects and etc. On top of that, with the integration of Augmented Reality (AR), visual searching and matching can now be done within seconds from the live camera view of the visual search application rather than taking photos and save as images before any matching to be done at the server end. However, existing technologies are still having limitations which stop MVS from performing as perfect as it aims to be. The physical appearance of an object, lighting conditions, angles of scanning are examples of the objective factors which has limit the performance of MVS. Hence, this led to the necessary of having an alternative way for users to perform a non-text searching when it is required. Subsequently, Quick Response (QR) code and 1D Barcode are chosen to fulfill the objective. In other words, the goal of this project is to integrate both QR code and 1D barcode scanning function into the existing application as well as update the search engine with new data in order for it to recognise barcode or QR code printed on a console game cover and redirect users to the Uniform Resource Locator (URL) of the game trailer. This will certainly allow users to have a better idea of the gameplay before any purchase and the application will also be able to switch to different scanning mode automatically upon user’s selection if the scanning inputs are detected to be invalid. At the end of this project, the search engine was updated with new data and the series of testing were conducted to ensure the integration of 1D barcode and QR code with the visual search are working seamlessly under one application. As a result, the application is now able to provide extra details of the console game after perform scanning on its cover photo using different types of scanning mode.
author2 Yap Kim Hui
author_facet Yap Kim Hui
Lok, Seng Siong
format Final Year Project
author Lok, Seng Siong
author_sort Lok, Seng Siong
title Visual product search in mobile commerce
title_short Visual product search in mobile commerce
title_full Visual product search in mobile commerce
title_fullStr Visual product search in mobile commerce
title_full_unstemmed Visual product search in mobile commerce
title_sort visual product search in mobile commerce
publishDate 2017
url http://hdl.handle.net/10356/72279
_version_ 1772828307305267200