NVIDIA Launches V2D Challenge to Teach Robots From Video
YunzhuLiYZ · x · 2026-09-17
NVIDIA Robotics has launched the Video to Data (V2D) Challenge, aimed at turning existing human videos into robot skills.
- Three tracks: 4D Reconstruction, Robotic Grounding, and Egocentric Video-to-Policy.
- Prizes: winning solutions announced at CoRL 2026; evaluation at scale; top teams invited to integrate code into a maintained repository; contributors can co-author the challenge's technical report.
- Judges include Yunzhu Li, Hao Su, Louis Lian and Changliu Liu.
- Registration closes September 21, 2026.
More from Embodied
- CMU Panel: Professor-Founders on Turning Robotics Research Into IPO-Scale Companies — ditzikow · 2026-09-17
- AGIBOT A3 Ultra humanoid enters real-world deployment in hotels, dealerships and metro — XxSpookxX · 2026-09-17
- Unitree G1 humanoid harvests okra outdoors: 70 days of dev, 16/20 success — CyberRobooo · 2026-09-17
- Sakana AI's Smart Cellular Bricks featured in Scientific American: 197 bricks self-identify and self-repair — SakanaAILabs · 2026-09-17
- Ex-Tsinghua/BIGAI team's generalist dexterity policy beats Astra on RoboDojo — chris_j_paxton · 2026-09-17
- GPT-Policy: In-Context Robot Learning with VLM Agents, No Gradient Updates — Dongzhou Cheng · 2026-09-17