Synthetic data for computer vision (CV) for training artificial intelligence algorithms is poised to gain widespread adoption by government and commercial customers to reduce cost and time of acquiring real sensor data
BELLEVUE, Wash., Dec. 15, 2022 /PRNewswire/ — Rendered.ai, the leading platform for physics-based synthetic data, wraps up 2022 by recapping its tremendous progress providing a commercial offering that empowers customers to design and generate customized datasets to train machine learning (ML) and artificial intelligence (AI) systems. Edge and remote sensing-based imagery collection are set to increase dramatically in 2023 and the need for tailored datasets to rapidly train and validate AI is going to drive significant demand for tools such as synthetic data.
“Demand from customers in diverse market segments such as life sciences, Earth observations, and manufacturing has demonstrated that CV Engineers and data scientists need access to better tools to create and customize datasets for AI training and validation,” said Nathan Kundtz, Ph. D., Founder and CEO of Rendered.ai. “Access to synthetic data as a capability through a subscription cloud offering has enabled our customers to explore synthetic data generation techniques to achieve initial AI performance benchmarks and then to revisit and improve training when business requirements change. We have already seen customers use synthetic data to achieve meaningful impact on real world outcomes, and we are proud to have developed the tools to get them there.”
Launched in January, Rendered.ai quickly established its first subscription customers by building on previous work in the synthetic satellite imagery space. Rendered.ai announced this Fall that its hosted platform for synthetic data generation is now available for subscription through the AWS Marketplace. Read about the benefits of subscribing to Rendered.ai’s Enterprise Subscription on the AWS Marketplace here.
In November, Rendered.ai announced expansion of its partnership with the Rochester Institute of Technology’s (RIT) Digital Imaging and Remote Sensing (DIRS) Laboratory providing broader access to the Rendered.ai remote-sensing synthetic data channel based on the Digital Imaging and Remote Sensing Image Generation (DIRSIG™) model. Read about how this collaboration bridges gaps within the scientific and government community here.
In December, the Hyperspectral Team at Planet Labs PBC announced collaboration with Rendered.ai to generate synthetic data for hyperspectral programs in advance of satellite launches to speed the effort to develop analytics and data delivery pipelines for customers. Learn more about the Planet-Rendered innovative collaboration here.
The Rendered.ai team is committed to engaging with communities who struggle with data availability and acquisition costs and complexity. For that reason, Rendered.ai joined the US Geospatial Intelligence Foundation and the Open Geospatial Consortium in 2022. Working with these organizations and leading technology partners such as NVIDIA, AWS, and Esri, Rendered.ai plans to reach more customers by connecting synthetic data to standards-driven data pipelines and workflows that are driving AI into critical decision-making systems.
Gartner estimates that by 2030, synthetic data will completely overshadow real data in AI models. Additionally, Gartner’s Innovation Insight for Synthetic Data from February 2022 stated, “We believe that synthetic data is important for the future of AI because it solves one of the most pervasive and critical challenges that AI systems face today — the lack of domain-specific, well-labeled, high-volume data at a reasonable cost.”
In 2023, Rendered.ai will be showcasing their platform as a service and customer projects at Esri’s Federal GIS Conference, the GeoBuiz Summit, the IEEE/CVF Computer Vision and Pattern Recognition Conference, and USGIF’s GEOINT Symposium.
If you’d like to hear about the latest news, tips, and product updates on synthetic data for computer vision applications sign up for Rendered.ai’s newsletter.
Rendered.ai is a Platform as a Service for synthetic data generation that puts physically accurate sensor modeling and a closed-loop data engineering workflow in the hands of data scientists and innovators. Founded by physicist Nathan Kundtz, Rendered.ai has created and powers the first-ever developer framework for synthetic data, turning simulation tools into synthetic data capabilities which includes scenario generation, 3D model libraries, asset management, compute management, annotation, metadata management and more. Rendered.ai is a privately held company based in Bellevue, Washington. For more information on the company and to sign up for a free account, please visit: www.rendered.ai.