May 26, 2022

The Israeli-born company said Datagen has raised a $50 million Series B round to accelerate the development of its synthetic data solutions for computer vision (CV) teams, bringing total funding to over $70 million. The company announced today. The round was led by new investor Scale Venture Partners, with partner Andy Vitus joining the DataGen board of directors.

With offices in Tel Aviv and New York, DataGen is building “a complete resume stack that will accelerate advances in AI by simulating real world environments to quickly train machine learning models at a much slower rate,” Vitus said. VC of Palo Alto predicts that “this will revolutionize the way we develop and test computer vision applications.”

Investors who invested $18.5 million in Series A Datagen 11 months ago entered this new round 11 months ago. This includes venture capital firms TLV Partners and Spider Capital, as well as Series A leader Viola Ventures, this time also through its growth division Viola Growth. AI and data dignitaries have also redoubled efforts, such as computer scientist Michael J. Black and Trevor Darrell, NVIDIA AI director Gal Chechik, and Anthony Goldblum, CEO of Kaggle.

Datagen CEO Ophir Zuk (Chakon) told gaming-updates that the list of investors could be long. Although the round closed a couple of weeks ago, the startup has left a “slightly delayed closure” and some names have yet to be confirmed.

When asked about Datagen’s most important post-A series milestone, Zuck explained that it had to do with building the self-service platform that target users requested in their original response. With this approach, DataGen now has a more scalable way to help customers self-generate the visual data they need to train their computer vision applications.

DataGen solutions are used by computer vision teams and machine learning engineers in a variety of organizations, including some Fortune 100 companies and large technology companies. It has a wide range of applications, but there are four that are growing faster than others, Zuck said: AR/VR/Metaverse, in the cockpit and cars in general, smart conferencing, and home security.

The car in the cab is a great example to better understand what DataGen does. The term refers to what happens in the car, such as whether the passenger is wearing a seatbelt or not. Passengers and cars come in many forms, and here AI comes to the rescue. Based on the first real 3D motion recordings, DataGen allows its customers to generate huge amounts of data to decide, for example, where to place an airbag.

We just touched on the common thread of synthetic data: how they take data from the real world and feed it into a data stream that needs more and more: abundant and rich, to remove biases, cover edge cases, etc.

DataGen focuses on visual data, but not on a specific industry. For example, as retail and robotics use cases begin to roll out, only specific real-life data needs to be collected, such as motion capture from warehouses. The algorithms and methods at the top are domain independent, Zuck said.

“The potential impact of what DataGen has to offer is huge for a wide range of applications,” Vitus said. Scale, a venture capital firm with over 20 years of history, has already invested in the Cognata car simulation platform and is optimistic about fake data. Like Zuck: “Synthetic data takes real data,” he summed up.

Leave a Reply

Your email address will not be published.