Atlanta, GA – Eureka Robotics today announced the launch of StereoPick, software for its AI Vision System that lets users set up, train and deploy 3D picking applications themselves.
Why vision projects take so long
3D vision is the hardest part of most robotic picking cells to get into production. Each new part typically needs a vision specialist to write custom code, capture and hand-label hundreds of training images, or tune CAD matching until it holds up under production conditions. A single application can take months, and because the know-how sits with the integrator or vision vendor, every new part starts the cycle again.
Putting 3D picking in users' hands
Eureka Robotics' goal is for 3D picking to be set up by the people who run the line, not outsourced for every new part. Its approach is to build the vision expertise into the software: an AI model that already knows how to find parts in a bin, training data generated from CAD instead of labeled photos, and a guided workflow in place of code. StereoPick brings these together in one tool.
What StereoPick is
StereoPick is the setup and training software of the Eureka AI Vision System. It takes an application from CAD file to live picks in six guided steps: users select the robot, upload CAD files for the gripper and the part, define where the part is picked and placed, and validate the cell on a digital twin. StereoPick then trains the vision model on synthetic images rendered from the CAD file and, once the hardware is in place, calibrates the camera to the robot. Simple, rigid parts typically train within a few hours, and everything, including training, runs on hardware in the plant without a cloud connection.
"With StereoPick, 3D picking becomes accessible to anyone. Either you bring a CAD file and the system trains itself, or you use the base model and start picking right away. The customer's own team is in control from setup to production," said Cuong Pham, CEO at Eureka Robotics.
More on StereoPick at eurekarobotics.com/eureka-3d-vision-software.
The Eureka AI Vision System
Eureka Robotics' AI Vision System is an integrated solution of Eureka 3D Camera, controller and software for bin picking and vision-guided robotics. A super-model trained on millions of parts, together with software accessible to entry-level users, reduces deployment time by months compared to conventional vision projects.
The system integrates with any robotic arm and operates as an open platform: models can run locally, users can train new parts in-house, and no subscription fees apply. The Eureka AI Vision System is deployed in production at Pratt & Whitney, Subaru and Coherent, and has completed more than 30 million picks.
Media contact: Julian Schaaf, VP Marketing, Eureka Robotics, julian@eurekarobotics.com
