Safetensors
haptics
vibrotactile
texture
flow-matching
encodec

HaptoFlow

HaptoFlow teaser

HaptoFlow is a Flow Matching vibrotactile generative model for interactive, real-time haptic rendering in VR. Conditioned on a material label and interaction parameters (stroking velocity and applied force), it generates 0.1 s waveform segments autoregressively, each conditioned on the previous one, so the vibration follows the user's motion. In the paper it outperforms all baselines in both waveform reproduction accuracy and inference latency, and its system latency stays well within the perceptual threshold of visual-haptic delay.

Paper | Project page | Code

Checkpoints

subfolder target signal
haptoflow_accel contact acceleration
haptoflow_audio contact sound

For details of the signals, see the Cluster Haptic Texture Dataset.

Citation

@misc{eguchi2026haptoflowhighfidelityrealtimevibrotactile,
      title={HaptoFlow: High-Fidelity Real-Time Vibrotactile Generation via Flow Matching for Virtual Reality},
      author={Michikuni Eguchi and Yuichi Hiroi and Takefumi Hiraki},
      year={2026},
      eprint={2608.01974},
      archivePrefix={arXiv},
      primaryClass={cs.HC},
      url={https://arxiv.org/abs/2608.01974},
}
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Dataset used to train tamago117/HaptoFlow

Paper for tamago117/HaptoFlow