Inspiration
The development and upkeep of roads to ensure road safety is difficult for countries that lack the infrastructure to support these large scale projects. Even the starting point of analysis and data collection has the potential to be streamlined using new technology.
What it does
I created a program that detects potholes using a dataset of multiple different arial and up-close views of roads.
How we built it
I used a combination of Google Colab, Jupyter Notebook, and Hugging Face to create the framework and provide the necessary data for object detection.
Challenges we ran into
There were a few limitations in the data, such as the accuracy of the bounding boxes in the training data, that affected the model's confidence.
Log in or sign up for Devpost to join the conversation.