AccuFit

Inspiration

In our team, most of members have had problem about maintaining weight. We found that the most common weight loss plan is to have a balanced diet and do exercise. Besides, we also had interviews with other colleagues about their experience about maintaining or losing weight. Through a few interview, we found all of them met the same problem that they lacked a user-friendly tool when recording their diet. They had to type nutrition facts by themself and it was quite inconvenient.

Therefore, our team came up with an amazing idea to simplify the processing of recording diet. We provides a tool that help people simply record daily diet by uploading food photo to the chatbot, and it will get the nutrition facts by recognizing patterns. Creating a user-friendly chatbot brings people a healthier life.

What it does

In the chatbot era, Our chatbot service named AccuFit provides a tool that can help people simply record their daily diet, making them build self-discipline in food intake. Our team applies pattern recognitions technique to complete this project. Users just upload food photo to the chatbot, AccuFit can get the nutrition facts of food by pattern recognition. Another tool is that people can record exercise and the system will calculate their daily remaining calories, recommending the healthy and suitable diet. Besides, people can also tap the button to open Google Maps and find where it is sold. Finally, the last tool is that a report will shows your daily intake of macronutrients (carbs, fats and protein) and calculate weekly change of weight.

How we built it

We use php combined with Microsoft's Custom Vision technology to achieve recognize patterns by using machine learning.

Challenges we ran into

One of the main challenges was to build the high-functional chatbot in a really short time(less than 2 weeks). Plus, our team members were all occupied with heavy workloads, we had limited time to discuss the project at the period of time. Therefore, we did a lot of research and planning individually, in order to elevate the efficiency when we could finally gathering.

Second, coming up with a real-time chart (the bar graph that shows daily remaining calories and calculate weekly change of weight) was also a challenge. Our development engineer did a lot of research and rewrote the program at least three times. Thankfully, through his perseverance and determination, we successfully tackled the obstacle.

What we learned

At the begging of this project, our team popped up with many ideas, but neither of them reach a consensus. Through continuous communication, we successfully blended those opinions into AccuFit. Along the way, we’ve improved our soft skills such as cross-functional communication and time management. As well as practical skills like pattern recognition and chatbot programming.