Inspiration

ÆON

Welcome to our repository! æon is a browser plugin that does the work of scouring through online videos for you. Simply visit a video page, activate the plugin, and see all of the video's highlights displayed visually for you, with nifty timestamps, the ability to easily jump to any of these key points in the video, and a (usually) precise estimate of where you'll find the original video thumbnail. Spend enough time on the video, and your viewing behaviour will update our servers with the intervals of video that you spent your time watching, helping others in the effort to sift through the online junk that just wastes everybody's time.

Tech

We use the Microsoft Azure platform to run our virtual machine as well as integrate our API smoothly into a Mongo database, which holds all of the relevant info about individual videos in nicely defined documents. Backend code is written entirely in python, and our beautiful user-interface is done in Javascript and makes use of a modified version of stock visualizations, which helps display viewing frequency.

Thumbnail analysis utilizes AI Computer Vision to compare the thumbnail image with frames in the video. It finds the most similar frame and indicates where the thumbnail is most likely to appear in the video.

We hope that you check out our plugin, and that you enjoy saving time with it as much as we do!

What it does

Aeon analyzes YouTube videos to give people a way to summarize YouTube videos, so that they can see the parts that are important to them.

How we built it

We used a MongoDB database to store our data, as well as a Python back end and a JavaScript front end, all hosted on Microsoft Azure.

Challenges we ran into

We had difficulties with some of the technologies that we weren't used to. For example, none of us were experienced with front end JavaScript, so we spent a large amount of time debugging and cleaning it up.

Accomplishments that we're proud of

Our thumbnail detection is quite accurate and a good example of basic computer vision.

What we learned

We learned that things are typically more difficult than you think! Also, sleep is pretty important for programmers. We learned how to use Microsoft Azure, as well as setting up a database, server, and frontend on our own.

What's next for Aeon

More intricate video parsing, for example using the video indexer on Microsoft Azure, could benefit the project as we could use some more complex data in our analysis.

Built With

Share this project:
×

Updates