SWIM: Compact Environment Representation for Reinforcement Learning Human Swimming

arXiv:2605.31120 · cs.GR, cs.AI, cs.LG · Submitted 2026-05-29 · Read on arXiv

cs.GR, cs.AI, cs.LG

Submitted: 2026-05-29

Updated: 2026-10-02

Code: https://github.com/InteractiveComputerGraphics/SPlisHSPlasH

Project page: https://binglunwang.github.io/SWIM

Terminology

• Reinforcement Learning — 1× in this paper · explained in 4 episodes

Sources

• MarineGym: A High-Performance Reinforcement Learning Platform for Underwater Robotics
• Smoothed Particle Hydrodynamics Techniques for the Physics Based Simulation of Fluids and Solids
• MimicKit: A Reinforcement Learning Framework for Motion Imitation and Control
• Proximal Policy Optimization Algorithms
• OceanSim: A GPU-Accelerated Underwater Robot Perception Simulation Framework
• Physics-based Deep Learning
• ControlVAE: Model-Based Learning of Generative Controllers for Physics-Based Characters
• MoConVQ: Unified Physics-Based Motion Control via Scalable Discrete Representations

Related papers

• SCRIPT: Scalable Diffusion Policy with Multi-stage Training for Language-driven Physics-Based Humanoid Control
• CADReasoner: Iterative Program Editing for CAD Reverse Engineering
• QuadLink: Autoregressive Quad-Dominant Mesh Generation via Point-Relation Learning
• DrawVideo: Grounded and Faithful Multi-Shot Video Generation from Storyboard Keyframe Sketches
• MotionPersona: Real-Time Locomotion Control across Personas, Bodies, and Styles
• GaussianBench: Physics-Fidelity Evaluation for Gaussian Scene Representations