SSI-Policy: Learning Structured Scene Interfaces for Vision-Language Robotic Manipulation
For robotic manipulation, this work addresses the challenge of learning spatial grounding and task-aware reasoning with limited data by providing a modular, RGB-only interface that decouples perception from control.
SSI-Policy introduces a Structured Scene Interface (SSI), an RGB-only intermediate representation that jointly encodes depth features, object layouts, and motion trajectories, enabling robot-agnostic policy learning from few demonstrations. On the LIBERO benchmark with only 10 demonstrations per task, it improves over the strongest prior method by nearly 15% and remains competitive with 50-demo methods using large-scale pretraining.
Real-world robotic manipulation demands spatial grounding, task-aware reasoning, and precise control. Learning such capabilities becomes particularly challenging in the low-data regime. Prior methods often trade off scalable task-level reasoning and explicit physical structure: video-based approaches can drift geometrically over long horizons, 3D approaches often require depth sensing, and many flow/trajectory interfaces emphasize motion without an explicit RGB-only geometric representation. We introduce SSI-Policy, a modular framework built around a Structured Scene Interface (SSI) -- a unified, RGB-only intermediate representation that jointly encodes monocular depth features, language-grounded object layouts, and instruction-conditioned 2D motion trajectories. Critically, SSI is robot-agnostic and trainable from action-free video, decoupling perception from control so that the downstream policy can learn from few demonstrations. On the LIBERO benchmark with only 10 demonstrations per task, SSI-Policy improves over the strongest prior method by nearly 15\% and remains competitive with 50-demo methods that leverage large-scale external pretraining. Ablations show that geometric and motion cues provide complementary benefits within the shared interface. We further validate on 13 real-world tasks spanning spatial reasoning, cross-embodiment transfer, and contact-rich manipulation.