Argon zero-shot generates robot sim environments from real egocentric video
keerthanpg · x · 2026-10-14
keerthanpg demos that Gemini 4 Argon can zero-shot generate sim environments from real video, including egocentric video — supporting both human egocentric to robot sim and robot egocentric to robot sim pipelines.
Combined with earlier posts, Argon's physical understanding also enables direct control of armed quadrupeds, real2sim2real improvement for full-body humanoids with dexterous hands, and hindsight labeling plus data quality control for robotics.
More from Embodied
- Google's Gemini 4 Argon for robotics: simulation training lifts task success from 56% to 75% — DynamicWebPaige · 2026-10-14
- Founding engineer leaves YC robotics startup to build his own robot from scratch — natesiggard · 2026-10-14
- "GPT-6 Astra" demo claims 25 sim environments, 25 tasks, 5 robot embodiments from 3 prompts in 24h — abhishekunique7 · 2026-10-14
- Dyna-2: million-hour world-action model lets robots learn tasks from 45 minutes of demos — Latent Space · 2026-10-14
- claude-as-cad: how to build real robots and parts without knowing CAD, by having Claude build its own CAD interface — _krishna_murthy · 2026-10-14
- Chronos Policy conditions robot control on real physical timestamps, not sequence indices — SongShuran · 2026-10-14