MOSAIC: Bridging the Sim-to-Real Gap in Generalist Humanoid Motion Tracking and Teleoperation with Rapid Residual Adaptation

• URL: http://arxiv.org/abs/2602.08594v2
• Date: Wed, 11 Feb 2026 16:45:16 GMT
• Title: MOSAIC: Bridging the Sim-to-Real Gap in Generalist Humanoid Motion Tracking and Teleoperation with Rapid Residual Adaptation
• Abstract summary: Generalist humanoid motion trackers have recently achieved strong simulation metrics by scaling data and training, yet often remain brittle on hardware during sustained teleoperation.<n>We present MOSAIC, an open-source, full-stack system for humanoid motion tracking and whole-body teleoperation across multiple interfaces.
• Score: 50.643488406438514
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Generalist humanoid motion trackers have recently achieved strong simulation metrics by scaling data and training, yet often remain brittle on hardware during sustained teleoperation due to interface- and dynamics-induced errors. We present MOSAIC, an open-source, full-stack system for humanoid motion tracking and whole-body teleoperation across multiple interfaces. MOSAIC first learns a teleoperation-oriented general motion tracker via RL on a multi-source motion bank with adaptive resampling and rewards that emphasize world-frame motion consistency, which is critical for mobile teleoperation. To bridge the sim-to-real interface gap without sacrificing generality, MOSAIC then performs rapid residual adaptation: an interface-specific policy is trained using minimal interface-specific data, and then distilled into the general tracker through an additive residual module, outperforming naive fine-tuning or continual learning. We validate MOSAIC with systematic ablations, out-of-distribution benchmarking, and real-robot experiments demonstrating robust offline motion replay and online long-horizon teleoperation under realistic latency and noise. Project page: baai-humanoid.github.io/MOSAIC.

Related papers

• X-OP: Cross-Morphology Whole-Body Teleoperation via MPC Retargeting [3.134050607080983]
  Whole-body teleoperation is essential for scalable robot data collection in loco-manipulation tasks.<n>We propose a hierarchical whole-body teleoperation framework driven by a single XR device.<n>We show higher success rates on whole-body control tasks for both a humanoid and a mobile manipulator.
  arXiv  Detail & Related papers  (2026-06-06T01:50:59Z)
• Real-Time Whole-Body Teleoperation of a Humanoid Robot Using IMU-Based Motion Capture with Sim2Sim and Sim2Real Validation [0.7325481380690131]
  This paper presents a real-time whole-body teleoperation system recorded with a Unitree G1 humanoid robot.<n>We introduce a custom motion-processing, kinematic, and control pipeline engineered for continuous, low-latency operation without any offline buffering or learning-based components.<n> Experimental results demonstrate stable, synchronized reproduction of a broad motion repertoire, including walking, standing, sitting, turning, bowing, and coordinated expressive full-body gestures.
  arXiv  Detail & Related papers  (2026-05-12T16:22:17Z)
• Interpretable Multimodal Gesture Recognition for Drone and Mobile Robot Teleoperation via Log-Likelihood Ratio Fusion [14.332919759770645]
  Vision-based gesture recognition has been explored as one method for hands-free teleoperation.<n>We propose a multimodal gesture recognition framework that integrates inertial data from Apple Watches on both wrists with capacitive sensing signals from custom gloves.<n>We show that our framework achieves performance comparable to a state-of-the-art vision-based baseline.
  arXiv  Detail & Related papers  (2026-02-27T05:52:04Z)
• PMG: Parameterized Motion Generator for Human-like Locomotion Control [14.637220434597168]
  We develop a real-time motion generator that produces human-like locomotion in a single integrated system.<n>We show that, within a single integrated system, PMG produces natural, human-like locomotion, responds precisely to high-dimensional control inputs.<n>These results establish a practical, experimentally validated pathway toward natural and deployable humanoid control.
  arXiv  Detail & Related papers  (2026-02-13T06:38:04Z)
• TeleGate: Whole-Body Humanoid Teleoperation via Gated Expert Selection with Motion Prior [27.930225667015872]
  This paper presents TeleGate, a unified whole-body teleoperation framework for humanoid robots.<n>Our key idea is to preserve the full capability of domain-specific expert policies by training a lightweight gating network.<n>To compensate for the absence of future reference trajectories in real-time teleoperation, we introduce a VAE-based motion prior module.
  arXiv  Detail & Related papers  (2026-02-10T10:14:06Z)
• TacUMI: A Multi-Modal Universal Manipulation Interface for Contact-Rich Tasks [35.05859151174601]
  We build on the idea of the handheld demonstration device Universal Manipulation Interface (UMI)<n>We introduce TacUMI, a multi-modal data collection system that integrates ViTac sensors, force-torque sensor, and pose tracker into a compact, robot-compatible gripper design.<n>We then propose a multi-modal segmentation framework that leverages temporal models to detect semantically meaningful event boundaries.
  arXiv  Detail & Related papers  (2026-01-21T00:14:28Z)
• VLNVerse: A Benchmark for Vision-Language Navigation with Versatile, Embodied, Realistic Simulation and Evaluation [61.82502719679122]
  We introduce VLNVerse, a benchmark for Versatile, Embodied, Realistic Simulation, and Evaluation.<n>VLNVerse redefines VLN as a scalable, full-stack embodied AI problem.<n>We propose a novel unified multi-task model capable of addressing all tasks within the benchmark.
  arXiv  Detail & Related papers  (2025-12-22T04:27:26Z)
• TSkel-Mamba: Temporal Dynamic Modeling via State Space Model for Human Skeleton-based Action Recognition [59.99922360648663]
  TSkel-Mamba is a hybrid Transformer-Mamba framework that effectively captures both spatial and temporal dynamics.<n>The MTI module employs multi-scale Cycle operators to capture cross-channel temporal interactions, a critical factor in action recognition.
  arXiv  Detail & Related papers  (2025-12-12T11:55:16Z)
• From Seeing to Experiencing: Scaling Navigation Foundation Models with Reinforcement Learning [59.88543114325153]
  We introduce the Seeing-to-Experiencing framework to scale the capability of navigation foundation models with reinforcement learning.<n>S2E combines the strengths of pre-training on videos and post-training through RL.<n>We establish a comprehensive end-to-end evaluation benchmark, NavBench-GS, built on photorealistic 3DGS reconstructions of real-world scenes.
  arXiv  Detail & Related papers  (2025-07-29T17:26:10Z)
• SETransformer: A Hybrid Attention-Based Architecture for Robust Human Activity Recognition [7.291558599547268]
  Human Activity Recognition (HAR) using wearable sensor data has become a central task in mobile computing, healthcare, and human-computer interaction.<n>We propose SETransformer, a hybrid deep neural architecture that combines Transformer-based temporal modeling with channel-wise squeeze-and-excitation (SE) attention and a learnable temporal attention pooling mechanism.<n>We evaluate SETransformer on the WISDM dataset and demonstrate that it significantly outperforms conventional models including LSTM, GRU, BiLSTM, and CNN baselines.
  arXiv  Detail & Related papers  (2025-05-25T23:39:34Z)
• Hybrid Neural-MPM for Interactive Fluid Simulations in Real-Time [57.30651532625017]
  We present a novel hybrid method that integrates numerical simulation, neural physics, and generative control.<n>Our system demonstrates robust performance across diverse 2D/3D scenarios, material types, and obstacle interactions.<n>We promise to release both models and data upon acceptance.
  arXiv  Detail & Related papers  (2025-05-25T01:27:18Z)
• Transformer Inertial Poser: Attention-based Real-time Human Motion Reconstruction from Sparse IMUs [79.72586714047199]
  We propose an attention-based deep learning method to reconstruct full-body motion from six IMU sensors in real-time. Our method achieves new state-of-the-art results both quantitatively and qualitatively, while being simple to implement and smaller in size.
  arXiv  Detail & Related papers  (2022-03-29T16:24:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.