[slaysians] Stretch925

Please take some time to check out our Github repro. If you like the idea, remember to like and click the VOTE button (on the right for web users and at the bottom for mobile users)!

Why Stretch925?

A small change, such as incorporating stretching into the work routine, can have a significant impact on employees' well-being and overall work experience. We believe that stretching can undo the harms of prolonged sitting, something which we have become widely accustomed to. Also, the mental escape and stress release can enhance clarity and productivity. By promoting this small change, we aim to create a positive and fulfilling work environment, where employees can experience the remarkable benefits of stretching and find joy in their work.

How it works

Stretch925 is an application that utilizes gamification and computer vision to promote a healthy stretching habit among corporate workers. An accessible, interactive, and smart digital solution, Stretch925 incentivises and rewards corporate workers for taking care of their well-being.

How we built it

To tackle the problem of mundane work, we decided to utilise gamification to bring out the fun among corporate workers. At the same time, since most corporate workers have some sort of camera devices, we also wanted to leverage on the use of computer vision or more specifically, pose estimation, to conveniently engage users and accurately detect actions performed by them. For the technicals, we used several technologies to build this project: Our web application is built using Next.js, which utilises Tensorflow.js for the inbuilt computer vision model. For the isolated computer vision model that is built for demonstration, we leveraged on PeekingDuck, a Python-based computer vision model, to obtain pose estimations for the detection of the stretch postures. To put the cherry on the cake, we also utilised Figma to conceptualize our mobile app and to protray how we would envision the eventual market product to be.

Challenges we ran into

Our team faced the most trouble finding the "eureka" moment during the initial ideation stage. Most of our ideas were either done before or simply lack the desired wow factor. We want the idea to be novel and also impactful enough to enhance user experience. Another factor to consider is that the solution needs to be extensive and provides marginal benefits to all stakeholders. This brings us to Stretch925 which is largely accessible, interactive, and cost-effective.

During the coding stage, we also had difficulties integrating the shell script and batch files from the computer vision program into Next.js. The team tried everything we could to resolve the issue and we concluded that security concerns over the data quality was the reason for the failed integration. It was heartbreaking because this was the final piece of the puzzle.

Accomplishments that we're proud of

A challenge is also an opportunity, the team certainly embraced this notion. We pivoted well towards our final idea and we couldn’t be more proud of one another. It was not easy to overcome the intellectual and emotional investment that we had put into our first idea. When we hit a wall with our first idea, we were not disheartened but were instead more intentional in identifying opportunities, we learnt to prioritise problems in sectors less optimised, problems less known, giving us the opportunity to innovate and wow our users. We eventually hit home with Stretch925.

Another accomplishment was that we never lost motivation no matter the issues we faced, period.

What we learnt

Even though all of us have some coding experience, it is always still an enriching experience rushing and working together to complete a functional prototype despite the tight deadline. From designing the prototype in Figma to presenting a working product in its respective languages and frameworks, it was a great journey honing our developing skills.

What's next for [slaysians] Stretch925

We believe Stretch925 will be the next revolutionary idea in bringing joy and practical benefits to the workplace, and that there are still many opportunities for Stretch925 to expand

As it is a computer vision project, there is still inaccuracy in the detection of the stretch postures. Moving forward, we will need to train our model and finetune it to be more efficient.
We also wish to collaborate with the Health Promotion Board and their ongoing initiatives to further incentivize users and encourage them to practice healthy stretching while they are at their desks.
Our team has bounced many fun ideas that could be considered for future implementation, such as team-based rankings and achievement badges!