Semi Automatic Video Annotation Tool

Project Title

Semi Automatic Video Annotation Tool

Project Area of Specialization Artificial IntelligenceProject Summary

Semi Automatic Video Annotation Tool will allow the user to annotate video on bases of the contents(objects) present in the video. These annotations can be used in content retrieval systems to quickly search and get the content.

Project Objectives

In this project, we have an intention to provide a desktop application that help users to annotate videos quickly and easily in an interactive way. Objective is to exploit the machine learning techniques to annotate video automatically and also gives user the authority to add its knowledge too to provide Semi Automatic experience.

Project Implementation Method

Tools and Techniques:

1. Matlab R2018a
2. Machine Learning using Neural Networks specifically called Supervised Learning.

Implementation:

Firstly, we will train our network on a dataset of specific domain. Our neural network will learn those concepts. After the learning/training process, user will be able to annotate his videos.

Benefits of the Project

• Help user to annotate video.
• Saves time
• Also enables the user to save the annotains for further use.

Technical Details of Final Deliverable

Desktop application that can train, save , load a neural network, can create a dataset and annotate videos.

Final Deliverable of the Project Software SystemType of Industry Media Technologies Artificial Intelligence(AI)Sustainable Development Goals Industry, Innovation and InfrastructureRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	80000
Printing	Miscellaneous	6	1000	6000
Stationary	Miscellaneous	6	500	3000
Overheads	Miscellaneous	4	250	1000
NVIDIA Graphic Card	Equipment	1	70000	70000