Hardware
PhysMamba: Selective State Space Models as Learned Articulated Body Simulators
arXiv:2610.04014v1 Announce Type: new Abstract: We introduce the first learned articulated body simulator based on a selective state space model (SSM), called PhysMamba. PhysMamba predicts next-frame
arXiv:2610.04014v1 Announce Type: new Abstract: We introduce the first learned articulated body simulator based on a selective state space model (SSM), called PhysMamba. PhysMamba predicts next-frame full-body state from position, rotation, and joint-action history, without velocity inputs. We compare four architectures under partial- and full-observation inputs and three training protocols. The from-scratch rollout training protocol gives Mamba2 strong short- and mid-horizon accuracy under partial observation (s10 = 43 mm, 2/50 diverged), while the two-stage teacher-based rollout protocol stabilizes GRU but fails for Mamba2. With CUDA graph compilation, Mamba2 reaches 0.107 ms per frame (9,334 FPS) on an H100 GPU, within 1.1imes of GRU's un-compiled throughput, adding under 1% latency to a 30 Hz HMR pipeline and enabling integration as a differentiable physics module for video-based mesh recovery.
Related
- Scal3R: Learning Efficient Multi-Relative Pose Query for Scalable Online 3D Reconstruction
- MultiGraspNet: A Multitask 3D Vision Model for Multi-gripper Robotic Grasping
- I2CD: Direct Image-to-Convex Decomposition for Simulation-Ready Collision Geometry
Source: arXiv cs.CV | 2026-10-06