Several times there is an issue that we liked some product of others and really don't know the name of the product to buy is for oneself, so taking this into account, we tried to build something which can make your search easier and faster.
What it does
Basically, it helps users to take the picture of the product they liked, and then give them the product description and also potential online services to order it.
How we built it
We generated a model using Machine Learning techniques like Tensorflow and convolutional neural networks and train the model on the dataset consisting of different product pictures and then created an app for user interaction and allows them to take a picture of what they liked.
Challenges we ran into
one of our major challenges was to train such higher image dataset. apart from this, we also train model several times to avoid overfitting.
Accomplishments that we're proud of
We able to build an initial prototype for detecting different watches at the moment but it can be trained with more images and will have larger labels at the end.
What we learned
We learned to work as a team and also shares our area expertise and could able to build an initial prototype with putting all together.
What's next for SLAAY
SLAAY can be scaled up for B2B so that it can be incorporated in the field of an automobile by the partners who are building parts for machinery and their customers can recognize the parts using our app and can place an order to them. It will help to reduce the maintenance time and increase customer satisfaction.