The transportation industry is undergoing major transformations that lead to increasing challenges in urban environments, such as traffic congestion, air pollution, crime, limited space, and inefficient waste and energy management.
Miovision’s mission is to address and resolve those urban challenges by providing cities with accurate data and tools to optimize traffic management, accessibility, road safety, parking management, and more.
Miovision collects multimodal traffic data using cameras, annotates them with V7, and then trains AI models to uncover actionable insights, helping municipalities create safer and more sustainable urban infrastructure.
Miovision leverages computer vision to monitor and understand patterns, fluctuations, traffic volumes, the efficiency of traffic networks, and ongoing infrastructure changes—in a fully automated way.
Visibility on metrics and annotators' work in V7 is very helpful to us, and it's something we didn’t have in our internal solution. The option to check past annotations and review the work is also valuable, along with V7’s ability to interactively define the workflows and the flexibility in task assignment.
Miovision builds computer vision-powered transportation solutions to improve mobility, increase safety and reduce congestion on city roadways. The company was founded in 2005 and has since worked with several municipalities, including Chicago, New Jersey, and Detroit.
The company currently offers three products: Miovision Scout (traffic data collection device), Miovision TrafficLink (remote traffic network management), and Miovision Traffop (signal performance measures tool).
Miovision collects visual data using cameras working 24/7 in different weather conditions to ensure that their datasets match the diversity of real-world conditions and capture edge case scenarios.
The company uses V7’s Auto-Annotation tool to label collected images with polygons to derive bounding boxes later (when needed).
Miovision imports data to V7 automatically using webhooks, organizes it into folders, auto-annotates it, and exports it to their database using another webhook to build training and validation sets and train their own neural networks.
Prior to discovering V7, Miovision developed and used their own internal solution for data labeling and training. However, due to high development and maintenance costs, the company decided to look for an advanced training data platform that would allow them to annotate a wider range of data and build robust workflows.
Furthermore, the Miovision team found V7’s interface easy to learn and navigate and the Auto-Annotate tool to be indispensable in helping them create training data faster.
We chose V7 because we wanted to build new types of workflows. We had our own system, but we wanted it to accomplish additional tasks, for example, create other annotations types, re-annotations, annotations on videos—activities that would be a lot of effort in development. V7 met our needs.
Thanks to V7, Miovision can annotate their data more efficiently and at a fraction of the cost. The team can also easily upload and manage their large-scale datasets, monitor annotation progress, build custom workflows, and deploy robust AI solutions faster.
In 2022, Miovision plans to build new workflows and annotate types of data they couldn’t annotate previously using their internal solution.
Discover how other AI-first companies solved knowledge tasks at scale with V7