• Shubham Tulsiani, CMU Robotics Institute assistant professor, receives NSF CAREER Award for groundbreaking perception research. • Award supports project bridging 2D video observations with 3D perception to enable robots to understand actions. • Research focuses on mapping human actions, like pressing a spray nozzle, to machine-understandable 3D models. • Goal: allow robots to generalize learned actions to new objects and environments in real-world settings. • Technical challenge: converting 2D internet videos into accurate 3D representations for action inference. • Tulsiani credits Ph.D. students for foundational work that led to the NSF proposal.

Article Summaries:

  • Carnegie Mellon University assistant professor Shubham Tulsiani has been awarded a National Science Foundation (NSF) CAREER Award for his research on perception systems that enable robots to better understand and act within the 3D world. The project seeks to bridge the gap between 2‑D observations-such as online videos-and 3‑D representations, allowing machines to infer how actions (e.g., pressing a spray nozzle) affect objects. By scaling this learning and translating perceptions into real‑world actions, Tulsiani’s work aims to improve robotic understanding of human‑performed tasks. The award, the NSF’s most prestigious recognition for early‑career faculty, acknowledges the foundational contributions of his Ph.D. students.

Sources: