Voxel51, a startup that develops a platform for analyzing unstructured data like images and videos, has raised $12.5 million in a Series A round led by Drive Capital, with participation from Top Harvest, Shasta Ventures, eLab Ventures and ID Ventures. Founder and CEO Jason Corso told TechCrunch that the new capital will be used to further grow the company’s platform and double the size of Voxel51’s team from 13 to 26 employees per year.
Corso, along with machine learning Ph.D. Brian Moore, says he created Voxel51 to harness the growing amount of unstructured data in AI and machine learning. Professor Corso at the University of Michigan says he sees a “critical need” for a better software infrastructure to support machine learning engineers and data scientists with visualization. Analyzing and understanding their data.
“Using unstructured and visual data is a huge challenge. Although we’ve seen recent successes in moving capabilities in ADAS from the lab to production, there are still challenges in bringing computer vision capabilities to production,” Corso told TechCrunch in an email interview.[Voxel51] Not for nothing. to get The right data to train a model, but the data that validates it will train your model to do what you want it to do.
Voxel51’s flagship product, FiftyOne, was launched as open source in August 2020. The tool aims to help developers visually analyze and optimize unstructured data sets for computer vision and machine learning use cases. This week marks the launch of a fully managed version of FiftyOneTeams that adds enterprise-focused features to the mix, including the ability to search image and video datasets stored in the cloud, share datasets with colleagues, and create and manage dataset versions. .
Corso says the open source version of FiftyOne — which supports on-premise, private and public cloud deployments — has more than 150,000 monthly active users, while the fully managed version has “half a dozen” Fortune 500 and “dozens” of startup paying customers. , namely in the automotive, retail, security and robotics industries.
We are completely open-source; New users have access to the full machine learning stack with unlimited data and usage. We don’t want to copy or transmit any media or metadata to our system,” Corso said. “Once deployed, FiftyOne becomes a trusted single source of truth for data across an enterprise.
Voxel51 sees competition from Aquarium Learning and Lightly.AI. But Corso doesn’t feel threatened. He said startups have thrived during the pandemic as the computer vision industry, which spans areas such as retail, security, advertising, agriculture and robotics, is booming, as well as unstructured demand for visual data and ways to manage it.
“Data is a key component of all machine learning processes; high-quality, clean data is a key component in building high-performance machine learning systems. However, the infrastructure and best practices for fully managing unstructured data are just the beginning,” said Corso. FiftyOne provides a window into an organization’s data processes and data quality, allowing them to deploy better machine learning systems and components if their data is ultimately unstructured.”
To date, Ann Arbor, Michigan-based Voxel51 has raised $15.8 million. Corso declined to disclose his earnings when asked.