-
-
Presentation
About Ramen Runners
As the adoption of Electric Vehicles and Autonomous Vehicles accelerates throughout the next few decades, the demand for vehicle detection through Computer Vision will inevitably increase. Companies such as Uber and Hertz, which have a fleet of their own vehicles will save a lot of time and money with an automated car model detection algorithm to keep track of inventory of cars.
We utilize two different models, CNN and ResNet-50, to classify different Toyota models given pictures of Toyota vehicles of different colors, at multiple different angles, and in a variety of backgrounds. Comparing the results of these models to identify car models is inspired by the Experiment I mentioned in the paper Deep Learning-Based Vehicle Make-Model Classification. We put our spin on it and chose to classify Toyota models because Toyota is by far the most popular car make and will continue to be in the United States.
Log in or sign up for Devpost to join the conversation.