While thinking about a problem to solve for this hackathon, we wanted to solve a problem that affected us. As photographers ourselves, we realized the importance of not only taking the best shot, but also selecting the best ones. We couldn't even begin to imagine how tedious and time consuming this job might be for someone that clicks thousands of photographs a day, like wedding photographers or marketing agencies. What if we could automate the boring stuff (no pun intended) and get AI to handle it for us? That's when Klar was born.

What it does

Klar uses Microsoft's Face API to help wedding photographers filter and sort through their wedding photos based on emotions that the people in the photo exhibit. This could be happiness, sadness, or (in the future), love and surprise. Photographers will also be able to tag a face and then get all the images of that person from their collection. We believe that this solution will save any photographer hours of time and effort.

How we built it

On the front-end, we wanted Klar to be as user-friendly and easy to use as possible. Klar is scalable. We recognize that not all photographers are technically-orientated when it comes to computers and AI, so we abstracted all of that from the end-user. Using HTML, CSS, and Javascript for animation, we built a beautiful but easy to understand front-end that anyone can use. The UX was designed to be uniform and match our logo so that it provides the end-user a seamless experience.

Challenges we ran into

Finding the right API to do emotion detection took most of the time during this hackathon, in addition to sorting out the usual bugs. We also had trouble selecting the right color scheme for our UX but then settled with pink, a mixture of red and white, which we thinks represents love and purity.

Accomplishments that we're proud of

For most of us, it was our first hackathon where we actually built something. Some of us entered with no coding experience and are now proficient in HTML and CSS. We're also proud of what we built. We strongly believe that something like Klar would be extremely useful for our target market because this is a real problem, and we have a real solution.

What we learned

Early on, we learned that a hackathon is about solving a problem, not about creating a cool solution (Thank you, Tank). So instead of implementing the cool, techy idea that we thought of on the way to the hackathon, we started from scratch and found an actual problem. In addition, apart from all the skills we acquired at this hackathon, working in a team was the most important thing that all of us learned.

What's next for Klar

For now, Klar is just a basic prototype that shows a fraction of the potential of a tool like this. Multiple features could be implemented that would make it more groundbreaking than it already is. We imagine implementing AI training models that would help Klar identify good advertisements (of food and products) from bad ones.

Built With

Share this project: