JUPITER usesgenAIto simulate realistically thebehaviorof deformable & soft objects to create corresponding computer vision models and grabbing strategies. It runs in all our robotic products, which comprises a framework to train AI models, a middleware to run a configurable pipeline including AI models, computer vision algorithms on an edge device, and a user interface for configuration, calibration, and - It is your mission to ensure low prediction failure rates under high generalization performance, as well as establishing standards for our in model training, experiment tracking and artifact generation. * 2-4 years work experience (computer vision applications (object detection, classification), vision language models (scene understanding)); * First contact with Vision Language
mehr