Research

Paper

TESTING February 25, 2026

WatchHand: Enabling Continuous Hand Pose Tracking On Off-the-Shelf Smartwatches

Authors

Jiwan Kim, Chi-Jung Lee, Hohurn Jung, Tianhong Catherine Yu, Ruidong Zhang, Ian Oakley, Cheng Zhang

Abstract

Tracking hand poses on wrist-wearables enables rich, expressive interactions, yet remains unavailable on commercial smartwatches, as prior implementations rely on external sensors or custom hardware, limiting their real-world applicability. To address this, we present WatchHand, the first continuous 3D hand pose tracking system implemented on off-the-shelf smartwatches using only their built-in speaker and microphone. WatchHand emits inaudible frequency-modulated continuous waves and captures their reflections from the hand. These acoustic signals are processed by a deep-learning model that estimates 3D hand poses for 20 finger joints. We evaluate WatchHand across diverse real-world conditions -- multiple smartwatch models, wearing-hands, body postures, noise conditions, pose-variation protocols -- and achieve a mean per-joint position error of 7.87 mm in cross-session tests with device remounting. Although performance drops for unseen users or gestures, the model adapts effectively with lightweight fine-tuning on small amounts of data. Overall, WatchHand lowers the barrier to smartwatch-based hand tracking by eliminating additional hardware while enabling robust, always-available interactions on millions of existing devices.

Metadata

arXiv ID: 2602.21610
Provider: ARXIV
Primary Category: cs.HC
Published: 2026-02-25
Fetched: 2026-02-26 05:00

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.21610v1</id>\n    <title>WatchHand: Enabling Continuous Hand Pose Tracking On Off-the-Shelf Smartwatches</title>\n    <updated>2026-02-25T06:12:43Z</updated>\n    <link href='https://arxiv.org/abs/2602.21610v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.21610v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Tracking hand poses on wrist-wearables enables rich, expressive interactions, yet remains unavailable on commercial smartwatches, as prior implementations rely on external sensors or custom hardware, limiting their real-world applicability. To address this, we present WatchHand, the first continuous 3D hand pose tracking system implemented on off-the-shelf smartwatches using only their built-in speaker and microphone. WatchHand emits inaudible frequency-modulated continuous waves and captures their reflections from the hand. These acoustic signals are processed by a deep-learning model that estimates 3D hand poses for 20 finger joints. We evaluate WatchHand across diverse real-world conditions -- multiple smartwatch models, wearing-hands, body postures, noise conditions, pose-variation protocols -- and achieve a mean per-joint position error of 7.87 mm in cross-session tests with device remounting. Although performance drops for unseen users or gestures, the model adapts effectively with lightweight fine-tuning on small amounts of data. Overall, WatchHand lowers the barrier to smartwatch-based hand tracking by eliminating additional hardware while enabling robust, always-available interactions on millions of existing devices.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.HC'/>\n    <published>2026-02-25T06:12:43Z</published>\n    <arxiv:comment>This work will be presented and published at ACM CHI 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.HC'/>\n    <author>\n      <name>Jiwan Kim</name>\n    </author>\n    <author>\n      <name>Chi-Jung Lee</name>\n    </author>\n    <author>\n      <name>Hohurn Jung</name>\n    </author>\n    <author>\n      <name>Tianhong Catherine Yu</name>\n    </author>\n    <author>\n      <name>Ruidong Zhang</name>\n    </author>\n    <author>\n      <name>Ian Oakley</name>\n    </author>\n    <author>\n      <name>Cheng Zhang</name>\n    </author>\n  </entry>"
}