Case Study FLIR: The global
sensor market

Working alongside FLIR’s engineering team as partners, CVEDIA designed a thermal synthetic pipeline that was able to train neural networks for a variety of use cases without the collection of more training data. CVEDIA integrated synthetic technology into FLIR’s engineering pipeline, providing FLIR with market-leading autonomous capabilities.

The pain points.

FLIR is a world-renowned company that designs and manufactures thermal cameras for both commercial and defense applications. The company's engineering team is focused on developing innovative products, but they faced a major obstacle in gathering enough training data for their computer vision algorithms. Despite researching various data collection methods, they realized that developing many of their use cases would be impossible without synthetic data due to the high data requirements and limited resources for extensive data collection. To address this issue, FLIR's CTO Pierre Boulanger explored the possibility of creating their own synthetic data pipeline. However, daunted by the complexity, he met with CVEDIA in 2018 and decided to invest in the company shortly after. With the help of CVEDIA, FLIR was able to enhance their AI roadmap and improve their products.

"Along the lines of creating and storing thousands of images of reel - it’s so expensive, even for a successful company like FLIR, it’s too daunting. We can’t do it with real images. We have to do it with synthetic images."

Pierre Boulanger
Chief Technology Officer

Our solution.

Over the course of a few years, CVEDIA collaborated with FLIR as their partner to develop innovative technology that had never been created before. This involved the creation of software that could train thermal neural networks for various use cases, solely using synthetic data. In early 2019, they were able to successfully solve the problem of domain adaptation. This synthetic data pipeline was then seamlessly integrated into FLIR's engineering pipeline, allowing CVEDIA technology to cater to a diverse range of applications, such as aerial, animal, and maritime.

As of 2021, FLIR has been spearheading the development of the next generation of ADAS and autonomous vehicle systems in the automotive community. Recently, the company launched a detection algorithm to prevent road collisions with deer, powered by CVEDIA's AI technology. In order to create a robust system, FLIR utilized CVEDIA's synthetic technology to supplement difficult-to-collect thermal deer data. Meanwhile, within the defense market, FLIR has employed CVEDIA technology to support several applications, including AiTR and troop protection. Given the challenges posed by sparse and often impossible data collection, synthetic algorithms have provided critical solutions to FLIR, enabling the company to maintain its position as a market leader in the smart sensor industry and offer substantial value to its clients, both in the commercial and defense sectors.

Start using CVEDIA-RT