Vehicle Classifier
The Vehicle Classifier is a deep learning-based algorithm designed to correctly interpret vehicle type including busses, pickups, sedans, bicycles, and more. Performing from both a car-mounted and elevated perspective, this algorithm is an asset for traffic data, security, and smart intersection applications, while conforming to a low power profile. It can be regularly updated with newly released vehicles according to your requirements.
PERFORMANCE
Optimized for edge performance with a low power profile.
RANGE
Up to 300 meters.
CONDITIONS
Dawn to dusk.
ENVIRONMENT
Environmentally agnostic performance.
ARCHITECTURE
Backbone: Resnet-101; Detector: RefineDet.
FUNCTIONALITY
Detection.
INPUT SIZE
512x512.
SENSOR MODALITY
Visible light.
Our services
MAINTENANCE
Each CVEDIA model comes backed with an ongoing maintenance agreement. Our team is available for continuous improvements and to ensure upkeep.
FINE GRAIN CLASSIFICATION
Speak to our team to discuss model classification as a service add on or as a separate model. CVEDIA’s in-house team will work with you to define your requirements, and build a synthetic-based model incorporating your classification specifications.
ADDITIONAL CLASSES
CVEDIA models are built using proprietary synthetic data technology, meaning adding additional classes to your model is possible in a matter of weeks.
Use cases
SMART CITIES
Use it as a reliable component of your smart intersection grid. The Vehicle Classifier performs with higher accuracy than traditional algorithms thanks to proprietary CVEDIA synthetic technology.
TRAFFIC AND DATA ANALYTICS
Calculate road usage analytics and highway traffic data with the ability to add new vehicle makes and models as they’re released.
SECURITY
Quickly localize vehicles of interest using security camera grids.
VEHICLE ADAS
Use it in your vehicle autonomy system to correctly classify vehicle size. This algorithm has been developed to maintain a low power profile, making it perfect for battery-powered vehicles and applications.