SimFoundry auto-builds interactive 3D sim scenes from video for robot policy training

DynamicWebPaige · x · 2026-11-10

DynamicWebPaige highlighted two robot-learning papers at the conference:

• SimFoundry (Fei-Fei Li's team, Poster S3 #81): reconstructs an interactive, sim-ready digital twin from a single real-world video, then generates unlimited affordance-preserving "digital cousins" across object, scene, and task axes for real-to-sim evaluation and sim-to-real policy training. The pipeline uses Gemini for scene layout and object articulation, plus Nano Banana for asset generation.
• Playful Agentic Robot Learning (Spotlight S6): lets a robot autonomously invent tasks, write control skills, and build a reusable skill library with Gemini, generating skills from browser interactions or real-world observation.

Original post →

More from Embodied

• Amazon Ring unveils $250 autonomous home security drone that patrols and streams live video — Polymarket · 2026-11-10
• NVIDIA-led study: unblocking VLM gradients in VLA training lifts manipulation progress from 10.3% to 32.9% — stepjamUK · 2026-11-10
• Google DeepMind launches Gemini Robotics 2.5 with 96% accuracy sorting Mercedes-Benz parts — GoogleDeepMind · 2026-11-10
• SlashVibe is a physical desk controller for managing multiple running AI coding agents — Scobleizer · 2026-11-10
• Path Robotics raises $300M Series E at $2.45B valuation to expand physical AI — Rewkang · 2026-11-10
• Unitree Go2 robot asks clarifying questions via Gemini and Bayesian active learning — DynamicWebPaige · 2026-11-10

Embodied channel →