Inspiration
From experiencing moments of forgetfulness to facing the challenges of dementia, the loss of cherished memories can bring unimaginable pain. For individuals with Alzheimer's and other conditions that affect memory, losing recollections of loved ones can feel like losing a part of their identity. We want to provide support for those experiencing memory loss, no matter how severe it is. Our goal is to offer a tool that can assist in memory recall and enhance cognitive abilities. Remindful is designed to complement existing dementia treatments (not replace them!!), providing a non-invasive, accessible, and enjoyable supplemental aid.
What do you remember about your birthday last year? Chances are it's the people, location, cake, or decorations. Key visuals such as these often stick out in our memories more and can help jolt lost memories.
What it does
Remindful makes use of active recall strategies and reminiscence therapy to help users recall their memories. It is both a health and educational tool that can help improve cognitive skills over time. It's a game that involves a picture depicting one of your memories. This could be a past event such as participating in UofTHacks!
Continuing with this example, imagine your hacking station right now. There are laptops sprawled across the table, sleeping bags on the floor, luggage set by your feet, and empty Red Bull cans waiting to be thrown out. But now, take away the laptops, the sleeping bags, the luggage, and the Red Bulls. This leaves you with an empty desk in the hacker room you chose on Friday.
Looking at this, try to remember what you were doing in here. Why were you in Myhal??
Users speak into the microphone, trying to recall this memory by making guesses and describing what they could have been doing.
"Hmmm I think I was working on a LAPTOP!"
The keyword "laptop" then prompts Remindful to reveal the laptops we had originally hidden.
Now the user continues to try to piece together the memory, while the picture slowly comes returns to its original, complete state.
We then end the game with a total score on how the user did, based on how many mistakes they made, if they were able to complete the image by remembering all the key objects , how long it took them, and other related factors!
How we built it
Our backend is built on Roboflow's YoloV8 object detection model which utilizies computer vision to detect the key objects (more on this in next steps) in the photo, python, and flask. Our frontend uses typescript and react to blur the images, build the user interface needed to make our web-app functioning. and handle the point system. We used the Web Speech API to transcribe your voice into text in real-time.
Challenges we ran into
- Cors ....
- Controlling blur to make sure the base image is still visible enough to make inferences
- Slow response times
- Connecting backend to frontend
Accomplishments that we're proud of
- We really tried to think outside the box for this year's theme and we think our project shows exactly that!
What we learned
Team was new to most of the technologies since our project idea was pretty niche and specific, but here are all the new technologies we learned!
- Roboflow's Yolov8 object detection model
- Image manipulation and blurring
- Real-time transcriptions and needing to work with expecting instant results
What's next for Remindful
- Build a model that determines the "importance" of an item such as frequency, identifiability, user feedback, etc...
- Optimize for faster response times
- More accurate object detection models
Log in or sign up for Devpost to join the conversation.