Hands-on, Hands-off: Gaze-Assisted Bimanual 3D Interaction

Hand Gesture RecognitionEye Tracking & Gaze InteractionMixed Reality Workspaces

Extended Reality (XR) systems with hand-tracking support direct manipulation of objects with both hands. A common interaction in this context is for the non-dominant hand (NDH) to orient an object for input by the dominant hand (DH). We explore bimanual interaction with gaze through three new modes of interaction where the input of the NDH, DH, or both hands is indirect based on Gaze+Pinch. These modes enable a new dynamic interplay between our hands, allowing flexible alternation between and pairing of complementary operations. Through applications, we demonstrate several use cases in the context of 3D modelling, where users exploit occlusion-free, low-effort, and fluid two-handed manipulation. To gain a deeper understanding of each mode, we present a user study on an asymmetric rotate-translate task. Most participants preferred indirect input with both hands for lower physical effort, without a penalty on user performance. Otherwise, they preferred modes where the NDH oriented the object directly, supporting preshaping of the hand, which is more challenging with indirect gestures. The insights gained are of relevance for the design of XR interfaces that aim to leverage eye and hand input in tandem.

Quick Actions

Share

Share this page

ios_share

https://hci.top/en/papers/uist/171002/2024

AdRecommended

Learn AI Coding at CodeNow

open_in_newOpen DOI Link
DOI: https://doi.org/10.1145/3654777.3676331
At a Glance

Paper Snapshot

fact_check
dataset
Source
UIST
calendar_month
Year
2024
emoji_events
Award
No award tagged
group
Authors
7 authors
sell
Subtopics
Hand Gesture Recognition, Eye Tracking & Gaze Interaction, Mixed Reality Workspaces
work
Professions
—
article
Content Status
Abstract only
hub
Related Papers
8 related papers