SimFoundry: Modular and Automated Scene Generation for Policy Learning and Evaluation
For robot learning researchers, SimFoundry automates scene generation for policy training and evaluation, reducing the need for costly real-world data collection.
SimFoundry is a modular system for zero-shot real-to-sim scene construction from video, enabling automated generation of digital cousins. Policies trained on SimFoundry data transfer zero-shot to real tasks, with simulation evaluations strongly predicting real-world performance (mean Pearson correlation 0.911) and task success rate improvements of up to 40% using task cousins.
Training and evaluating robot policies in the real world is costly and difficult to scale. We introduce SimFoundry, a modular and automated system for zero-shot real-to-sim scene construction from a video. SimFoundry generates sim-ready digital twins and supports object, scene, and task editing, enabling the automated generation of diverse digital cousins: affordance-preserving variations of reconstructed real-world scenes. Policies trained on SimFoundry data transfer zero-shot to challenging real tasks involving multi-step manipulation, articulated object interaction, and bimanual interaction, and its digital cousins (variations of the original scene, objects, and tasks) facilitate generalization to new real-world conditions. Across 7 manipulation tasks and 5 policy architectures, SimFoundry simulation evaluations strongly predict real-world performance, with mean Pearson correlation 0.911 and mean maximum ranking violation 0.018. When evaluating sim-trained policies zero-shot in the real world, policies trained with object, scene, and task cousins in simulation show average task success rate improvements of 17%, 21%, and 40%, respectively. Additional details at https://research.nvidia.com/labs/gear/simfoundry/ .