PALO ALTO, Calif., May 5, 2021 /PRNewswire-PRWeb/ -- OpenCV, the world leader in Computer Vision, and Roboflow, an end-to-end computer vision developer platform, have launched the most streamlined deployment pipeline for OpenCV AI Kit (OAK).
OAK, developed in partnership with Luxonis, is the world's first spatial AI and CV camera and has a global user base of tens of thousands and a record-breaking $1.35M Kickstarter campaign.
Roboflow's new features allow developers to train, refine, and deploy models to OAK devices — without being machine learning experts or reinventing the wheel on machine learning operations. With this update, OAK moves to the head of the table alongside best-in-class Roboflow-supported deployment options like the NVIDIA Jetson.
Developers are now able to train a custom model and deploy it via a Docker container directly to their OAK devices without wrestling with dependency issues or converting between model formats and intermediaries. This new functionality broadens the usability of artificial intelligence models and devices, making it easier than ever for developers to rapidly iterate and improve on their datasets and neural models.
Satya Mallick, CEO of OpenCV said: "Our mission is to serve a large community of AI developers by creating performant hardware with easy to use developer tools. To bring the best to our community, we work with the best. That is why we have partnered with Roboflow and Luxonis to make the AI pipeline ridiculously simple for programmers with limited prior experience in AI."
Roboflow, an OpenCV Silver Partner, will continue to support the OAK ecosystem with exclusive Courses, blog posts, and other collaborations. Roboflow CEO Joseph Nelson recently appeared on the OpenCV Weekly Webinar discussing "Why Computer Vision Projects Fail" and engaging in a Q&A session with OpenCV community members in attendance.
Existing owners and new purchasers of OpenCV AI Kit, including backers of last summer's successful Kickstarter campaign, receive free training credit on the Roboflow platform, further showing Roboflow's support for the global OpenCV community.
Developers can get started with this new feature at https://roboflow.com/opencv-oak
Luxonis CEO Brandon Gilles, said: "At Luxonis, we're excited about the Roboflow platform as it allows our users to manage and understand their datasets plus re-train their models quickly and easily. And with Roboflow's direct integration with the DepthAI OAK devices, these datasets and models can be field-refined and continuously improved over time. Coupled with Luxonis' Spatial AI and CV ecosystem, this allows for low-risk product development and production-grade solutions."
Joseph Nelson, Roboflow CEO, said: "Luxonis is enabling a new wave of embedded computer vision capabilities. With Roboflow's tools, users can collect data from their DepthAI OAK devices via API, annotate, and seamlessly train and deploy a model to their OAK. With the power of Roboflow and OAK, we've seen users build products from medical inventory scanners to 'Roombas for dog poop.' It's a creative community of builders we're excited to support."
Roboflow is a technology company that enables developers to use computer vision even without machine learning expertise. Over 30,000 users from businesses of every size — from startups to public companies — use the company's end-to-end platform for image and video collection, organization, annotation, preprocessing, model training, and deployment. Roboflow provides the tools for companies to improve their datasets and build more accurate computer vision models faster so their teams can focus on their domain problems without reinventing the wheel on vision infrastructure. Over the long term, Roboflow is building the operating system for visual computing.
Luxonis is on a mission to improve the engineering efficiency of embedding performant, spatial AI and CV into actual products. Luxonis' DepthAI ecosystem of open-source hardware, firmware, and software allows engineers to proof-of-concept in a day and get a product-like prototype in a month. Permissively-licensed, DepthAI can be integrated to closed-source and open-source projects and products alike.
OpenCV.org is a non-profit organization committed to serving a large and growing AI community by building an ecosystem of AI products and services. In addition to its flagship library, OpenCV.org and its partners create courses, design hardware, and provide consulting services for AI. OpenCV.org supports and informs the community through its forum and newsletter.
Satya Mallick, OpenCV, 813-541-4342, [email protected]