Most artificial intelligence models are trained through supervised learning, meaning that humans must label raw data. Data labeling is a critical part of automating artificial intelligence and machine learning models, but at the same time, it can be time-consuming and tedious work.
A Korean startup called AIMMO, which uses software and humans to label and categorize image, video, sound, text and sensor fusion data, built an AI data annotation platform, making the data labeling faster for enterprises.
AIMMO announced today it has raised $12 million in a Series A round to advance its data labeling technology and spur global expansion. Seven venture capital firms participated in the latest round: DS Asset Management, Industrial Bank of Korea, Hanwha Investment & Securities, S&S Investment, Toss Investment, Korea Asset Investment & Securities and Venture Field. AIMMO declined to comment on its valuation.
“The pandemic accelerated transforming contactless technology with AI deployment for intelligence surveillance, smart city, autonomous driving, smart factory and robotics where AI data is inevitably needed,” Doyle Chung, head of Global Sales of AIMMO, said in an email interview. “Among all the directions and industries, our focuses are [mainly] smart city and autonomous driving.”
Founded in 2016 by chief executive officer SeungTaek Oh, the startup has three data annotation tools: AIMMO DaaS, which manages sensor fusion data for autonomous vehicle corporations; AIMMO GtaaS, a turnkey-based platform for big data; and AIMMO Enterprises, launched in 2020, a web-based SaaS annotation labeling tool using cloud architecture.
The startup said its tools help improve the inefficiency of the data annotation process so that customers focus only on their AI modeling. There’s no platform fee, coding skills or installation required for AIMMO Enterprises, with which users can label their data via browsers like Chrome. As for its AIMMO GtaaS, users send raw data to AIMMO that labels the data with inspection and delivers it back to the customers, Chung said.
Through the AIMMO DaaS platform, its sales and volume of its data labeling increased by 200% in 2021 compared to the previous year. AIMMO posted $10 million of revenue in 2021, according to its IR pitch deck. The company expects growth in 2022, with the global demand for the autonomous driving field continually increasing.
ScreenShot | AIMMO website
The global data collection and labeling market size was valued at $1.6 billion in 2021 and is projected to reach $8.2 billion in 2028, according to a market analysis report by Grand View Research.
AIMMO works with an array of companies, including carmaker Hyundai Motor, car parts developer Hyundai Mobis, car-hailing startup Kakao Mobility, car-sharing startup SoCar and autonomous cargo developer ThoreDrive. AIMMO also works with non-autonomous vehicle companies in robotics, optical character recognition (OCR), smart factory, intelligent surveillance, e-commerce and logistics industries like telco corporation SK Telecom, internet giants Naver and Kakao and Japanese construction equipment maker.
The Korea-headquartered startup has offices in the U.K, the U.S., Japan and Vietnam. Chung said it also plans to open offices in Germany and Canada in 2022. AIMMO will compete with companies like Scale AI, Playment, Understand.ai and Deepen AI when it expands further to the global market. Today, the startup has 200 employees across the globe and more than 10,000 data labelers.