[TPAMI2026 / RSS2025] Code for my paper "You Only Teach Once: Learn One-Shot Bimanual Robotic Manipulation from Video Demonstrations"
Official implementation of VGGT-DP: Generalizable Robot Control via Vision Foundation Models (arXiv:2509.18778). A diffusion policy that uses the frozen VGGT vision foundation model as a geometry-aware visual encoder for robot manipulation, evaluated on MetaWorld.
Vision-only goal-conditioned quadrotor navigation in 3D Gaussian Splats: MPC distillation via BC, DAgger, and offline-to-online RL (SAC, residual TD3+BC)
First native Apple Silicon (MLX) port of Diffusion Policy (RSS 2023 Best Paper). 6 policy variants, 472 tests, Metal GPU verified. Train and run visuomotor diffusion policies on M-series — no CUDA required.
A modular ROS 2 framework for training and deploying Generative Visuomotor Policies (Diffusion, Flow Matching) for hardware-agnostic robotic manipulation.
To associate your repository with the visuomotor-policy topic, visit your repo's landing page and select "manage topics."