Visual Information Retrieval

INDPENDENT STUDY COURSE

Mariam Rashed Saif Humaid Hedhaiba Saif Al Dhaheri

Topic:

Visual Information Retrieval

Visual search engines for automatic information retrieval and classification are already available for content-based web browsing/analysis, e.g.

[url removed, login to view]

[url removed, login to view]

[url removed, login to view]

[url removed, login to view] (Chinese)

[url removed, login to view] (Chinese)

This is a very promising area for the future internet (including the internet of things).

The objective of this study is to familiarize with the theoretical fundamentals and practical results of visual information retrieval.

In particular, the student should understand principles of algorithms and techniques used in visual information retrieval, and to evaluate performances of publicly available visual search engines.

Mode of study:

a) Literature study and internet search on visual information retrieval: theory, practical issues and applications;

b) Analysis of algorithmic mechanisms used in the visual search engines;

c) Performance evaluation of available visual search engines;

d) Summary of the acquired knowledge and experiences in the reports.

Deliverables and milestones (assessment):

a) Familiarization with fundamentals of information retrieval (visual information retrieval, in particular). Exemplary starting points:

• “Introduction to Search Engine Technology” course (available at [url removed, login to view])

• Wikipedia page on image retrieval [url removed, login to view]

A survey of typical algorithms and methods

b) Familiarization with the most popular keypoint-based techniques prospectively used in the content-based visual information retrieval. Simple programming experiments to be conducted as a supplement providing a better understanding of the techniques. The literature will be supplied by the lecturer. The results to be included in the summary in point (c) and in the report in point (d). ( I have the book )

c) Experimental verification of performances and comparative analysis of the search engines for visual information retrieval. In particular, text-based retrieval should be compared to content-based retrieval. Summary of the experimental analysis to be submitted Alternatively; the summary can be submitted with the final report. (I have the software)

Research Technical Writing

Project ID: #6792782

About the project