PeriphAR: Fast and Accurate Real-World Object Selection with Peripheral Augmented Reality Displays

Abstract

Gaze-based selection in XR requires visual confirmation due to eye-tracking limitations and target ambiguity in 3D contexts. Current designs for wide-FOV displays use world-locked, central overlays, which are not conducive to always-on AR glasses. This paper introduces PeriphAR (per-ree-far), a visualization technique that leverages peripheral vision for feedback during gaze-based selection on a monocular AR display. In a first user study, we isolated text, color, and shape properties of target objects to compare peripheral selection cues. Peripheral vision was more sensitive to color than shape, but this sensitivity rapidly declined at lower contrast. To preserve preattentive processing of color, we developed two strategies to enhance color in users' peripheral vision. In a second user study, our strategy that maximized contrast of the target to the neighboring object with the most similar color was subjectively preferred. As proof of concept, we implemented PeriphAR in an end-to-end system to test performance with real-world object detection.

Figures (17)

• Figure 1: Our system consists of four components: (1) peripheral AR display emulation simulates monocular AR glasses with limited FOV on Quest Pro, while (2) gaze-driven object selection uses eye tracking and video passthrough for object selection; (3) real-world target simulation enables simulation of target objects using virtual 3D shapes and models in world space, while (4) peripheral proxy generation creates a visual representation of the real-world target on the simulated display screen. The system was used in two studies to determine key object properties for visually salient target representations.
• Figure 2: Our system measures display peripherality based on gaze dynamics: a design that requires less time to look at, and fewer gaze transitions into, the display means it provides better peripherality. By structuring the visual field into a gradient (blue) around the simulated display, we compute the ambient value ($A$) of each gaze sample as: (1)$A=0$, when the vector points in-world and away from the display, (2)$0<A<1$, when in the blue zone modulated within 10$^{\circ}$ in each direction around the display, or (3)$A=1$, when in-display and intersecting with the simulated display's screen area (dotted blue box).
• Figure 3: The tetris game in Study 1 simulated real-world targets using virtual objects in the shape of tetrominoes in four display conditions: the snapshot condition combined the text label (O), square shape in color (blue O), and surrounding objects (green T). The text, color, and shape conditions isolated these attributes as shown on the right.
• Figure 3: Objective metrics.
• Figure 4: The four display conditions used in the tetris game: snapshot with the text label and image of the target with surrounding shapes; text with only the text label; color with only a circle filled in the target's color; shape with only the target's shape isolated in gray.