Optimizing Curve-Based Selection with On-Body Surfaces in Virtual Environments

TL;DR

This work addresses the challenge of precise object selection in dense VR environments by combining a finger-gesture–driven quadratic Bézier Curve with on-body forearm projection to improve occlusion handling and immersion. It introduces a Bezier-curve generator controlled by finger curvature $\kappa$ and a real-time proximity-matching projection that maps the four nearest objects onto the forearm, implemented via a 20-segment discretization and real-time projection using $\text{Vector3.Project}$. A preliminary qualitative study with $N=24$ compares four conditions (Mid-air/On-body × Linear Ray/Bezier Curve), revealing that Bezier curves excel in occluded/dense scenes while on-body projection enhances immersion, albeit with precision and fatigue trade-offs. The work highlights a promising direction for dense VR interfaces, with future work focusing on improved proximity-matching accuracy, quantitative performance metrics, and broader scene configurations.

Abstract

Virtual Reality (VR) interfaces often rely on linear ray-casting for object selection but struggle with precision in dense or occluded environments. This late-breaking work introduces an optimized dual-layered selection mechanism combining dynamic Bezier Curves, controlled via finger gestures, with on-body interaction surfaces to enhance precision and immersion. Bezier Curves offer fine-grained control and flexibility in complex scenarios, while on-body surfaces project nearby virtual objects onto the user's forearm, leveraging proprioception and tactile feedback. A preliminary qualitative study ($N$ = 24) compared two interaction paradigms (Bezier Curve vs. Linear Ray) and two interaction media (On-body vs. Mid-air). Participants praised the Bezier Curve's ability to target occluded objects but noted the physical demand. On-body interactions were favored for their immersive qualities, while mid-air interactions were appreciated for maintaining focus on the virtual scene. These findings highlight the importance of balancing ease of learning and precise control when designing VR selection techniques, opening avenues for further exploration of curve-based and on-body interactions in dense virtual environments.

Figures (1)

• Figure 1: (1) Generation of a Bézier Curve defined by start point $P_0$, control point $P_1$, and end point $P_2$, using parameters such as curvature $\kappa$ and length $\ell$. (2) Example of how the proximity-matching mechanism works, demonstrating the calculation of the shortest distance from an object $O$ to points on the curve, specifically between $S_i$ and $S_{i+1}$, and the projection $O_{proj}$. (3) A schematic representation of using the on-body Bézier Curve with the proximity-matching mechanism, where the closest four objects are selected and projected onto the user's forearm.