Efficient and Accurate Object 3D Selection With Eye Tracking-Based Progressive Refinement

Abstract

Selection by progressive refinement allows the accurate acquisition of targets with small visual sizes while keeping the required precision of the task low. Using the eyes as a means to perform 3D selections is naturally hindered by the low accuracy of eye movements. To account for this low accuracy, we propose to use the concept of progressive refinement to allow accurate 3D selection. We designed a novel eye tracking selection technique with progressive refinement–Eye-controlled Sphere-casting refined by QUAD-menu (EyeSQUAD). We propose an approximation method to stabilize the calculated point-of-regard and a space partitioning method to improve computation. We evaluated the performance of EyeSQUAD in comparison to two previous selection techniques–ray-casting and SQUAD–under different target size and distractor density conditions. Results show that EyeSQUAD outperforms previous eye tracking-based selection techniques, is more accurate and can achieve similar selection speed as ray-casting, and is less accurate and slower than SQUAD. We discuss implications of designing eye tracking-based progressive refinement interaction techniques and provide a potential solution for multimodal user interfaces with eye tracking.