Project Proposal: Crowdsourcing Web Portal for Video Quality Data Collection

Background:

Image and Vision Computing(IVC) Lab lead by Prof. Zhou Wang in ECE Department is working on a vision to make better perceptual video and image quality assessment models. To succeed in their vision, lab wants to create a huge video quality feedback database through crowdsourcing. Our project’s aim is to develop web portal for the lab to collect this data more efficiently and at reduced cost.

Scope:

• To create UI for subjective human experiment(showing videos and collecting feedback)
• To create backend module interacting with frontend, database(videos and feedback score) and score aggregator module.
• Implement an active learning based algorithm to select a set of videos(~10 videos) from a huge database of videos(~Million) in such a way that least number of experiments will be required to gather maximum information.

### Objectives In this project we would like to achieve the following objectives:

• Understand the Agile Software development process by adopting the underlying principles in the project.
• Get familiar with Full-Stack Web Development, which includes learning about different technologies and frameworks used in the making of a website.
• Specifically, we would like to work with the following technology stacks:
  • Front End: HTML/CSS, JavaScript Bootstraps
  • Backend: Django/Postgresql / Amazon S3
  • Project Management Tools: GitLab

Timeframe:

Phases	Tasks	Start Date-End Date
Phase One	Build a basic functional prototype while learning web development technologies stack.	14-Jan-2019 to 11-Feb-2019
Phase Two	Implement Specific functionalities which will keep on evolving.	12-Feb-2019 to 4-Mar-2019
Phase Three	Website Deployment, Final Testing and Client’s Approval	4-Mar-2019 to 31-Mar-2019

Key Stakeholders

Client	UW IVC Lab(Wentao Liu - PhD Student)
Team	Anirudh, Ganesh, Jatin, Navni, Vaishali, Zeba