My first inspiration for this project came from the following reddit post: Fellowship of the Ring “movie barcode”, a slice of every single frame from the movie.

Made famous by the following Tumblr blog, “moviebarcodes” is a technique developed to compress an entire movie into a single image. The result of this technique is a single bar consisting of generated colors for every frame of the movie. In some moviebarcodes, narrative elements can be distinguished for the movies with the usage of color theory. Applications of moviebarcodes range from aesthetically pleasing posters to a tool for Disney animators.

In short, developers are taking entire movies, condensing them, and representing every frame as a color value found for that time. They can then create a color palette from the generated moviebarcode. What I find so interesting about moviebarcodes is that over two hours of video content can quickly be “digested” by a user in a single glance.

After seeing moviebarcodes in action, I began considering possible applications to my work with COSMOS. One of the major programs I work with, YoutubeTracker, is a research tool for Youtube Analysis that is currently in development by COSMOS at UA Little Rock. The goal of YouTubeTracker is to make finding insights for Youtube accessible to people that are not data scientists but are interested in researching a particular topic.

For my project idea, I hypothesize that YoutubeTracker can utilize moviebarcodes to effectively visualize data and provide impactful insights. Based on the feedback I have received in COSMOS and my social computing class, I feel confident taking this idea forward as the official topic for my project. My next step now is to secure the members of my project committee.