mmPencil: Toward Writing-Style-Independent In-Air Handwriting Recognition via mmWave Radar and Large Vision-Language Model

Key points

• We collect and release a 3D mmWave handwriting dataset containing 7,664 samples of 200 words from 12 users across 22 writing scenarios.1
• Extensive experiments demonstrate that mmPencil achieves robust word recognition under variations in writing position, range, angle, speed, size, direction, and user movement.1
• It achieves 97.60% accuracy on 200 word classes for seen users and 92.50% zero-shot accuracy on 50 word classes for unseen users, outperforming state-of-the-art baselines.1

Sources (1)

• [1]mmPencil: Toward Writing-Style-Independent In-Air Handwriting Recognition via mmWave Radar and Large Vision-Language Model
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 16, 01:08 PM

  In-air handwriting recognition based on radio frequency (RF) signals offers a promising solution for human-computer interaction, yet existing methods are often limited by predefined writing planes, constrained writing styles, and small recognition vocabularies.

  We collect and release a 3D mmWave handwriting dataset containing 7,664 samples of 200 words from 12 users across 22 writing scenarios.

Extractive summary: sentences quoted from the sources.