The primary model used for the creation of our AI product is YOLO. YOLO mainly uses Convolutional Neural Networks (CNN), which are a special type of neural network based on the convolution operator. The model outputs bounding boxes (like the yellow rectangle which can be seen in the image on the left) as well as their confidences and class probabilities (an example of a class is cellphones).
AI Camp Breakdown
All the AI Camp students were welcomed, and although the camp was going to require rigorous efforts of Python programming, some students had never coded. So, over the course of 3 to 4 days, the students were tasked with learning this language. Then, the collection of
data began, and we also learned about graphing with matplotlib, which would help us with numpy and pandas in Week 2.
As the camp progressed, we learned about numpy and pandas. These two modules are essential in graph creating, and plotting data. Towards the end of the week, we worked with the IOU module, along with using said module to create confusion matrices. These matrices were then used to identify objects based on their classification. Moreover, we ran several tests to find the precision and recall of our model (both of which were quite high).
Eventually, AI Camp reached its final week. To prevent the software from crashing, bug fixes were made. Now that the program was at peak performance, it was time to work on the website. Some members designed the visuals while others worked on programming the hyperlinks to different parts of the site. At the end of the week, we finalized the website and project and delivered our presentation.