Track-and-Complete: Learning Humanoid Skills from a Single Failed Human Video
cs.RO, cs.CV
Submitted: 2026-09-29
Updated: 2026-09-29
Project page: https://tracc-humanoid.github.io
Terminology
Sources
- MeshMimic: Geometry-Aware Humanoid Motion Learning through 3D Scene Reconstruction
- HDMI: Learning Interactive Humanoid Whole-Body Control from Human Videos
- LUCID: Learning Embodiment-Agnostic Intent Models from Unstructured Human Videos for Scalable Dexterous Robot Skill Acquisition
- RoboMirror: Understand Before You Imitate for Video to Humanoid Locomotion
- Imitation Learning from Imperfect Demonstration
- Extrapolating Beyond Suboptimal Demonstrations via Inverse Reinforcement Learning from Observations
- Text2Reward: Reward Shaping with Language Models for Reinforcement Learning
- DrEureka: Language Model Guided Sim-To-Real Transfer
- Vision-Language Models are Zero-Shot Reward Models for Reinforcement Learning
- OpenAI GPT-5 System Card
- TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization
- Proximal Policy Optimization Algorithms
Related papers
- FMT x: An Efficient and Asymptotically Optimal Extension of the Fast Marching Tree for Dynamic Replanning
- MPCFormer: A physics-informed data-driven approach for explainable socially-aware autonomous driving
- RoboLab: A High-Fidelity Simulation Benchmark for Analysis of Task Generalist Policies
- HRDexDB: A 4D Dexterous Grasping Dataset Across Human and Multiple Robot Embodiments
- APT: Action Expert Pretraining Improves Instruction Generalization of Vision-Language-Action Policies
- Fine-tuning is Not Enough: A Parallel Framework for Collaborative Imitation and Reinforcement Learning in End-to-end Autonomous Driving