*Result*: Human–Machine Coupling Study for Fast Visual Target Capture.
*Further Information*
*With the continuous advancement of computer vision, human–computer interaction, and intelligent sensing technologies, accurate line-of-sight (LOS) tracking has become a critical research topic. However, traditional gaze tracking systems face significant limitations under conditions of natural head movement. The direction of human vision is determined jointly by head posture and eye movement, and precise acquisition of head pose remains a key challenge in gaze tracking technology. To address the limitations imposed by head movement, this study aims to improve the robustness of gaze tracking in natural interaction scenarios. This paper proposes a novel head pose estimation method based on the 3D spatial information of facial feature points. The method utilizes stereo vision to reconstruct 3D feature points and employs a geometric model to calculate head pose, effectively decoupling head movement from eye movement in gaze tracking. In the experimental setup, two interaction modalities were tested: head pointing combined with key pressing, and gaze estimation combined with head pointing. Experimental results demonstrate that the proposed method achieves interaction latency between 100 ms and 200 ms, with a fast capture success rate of up to 85%, slightly outperforming traditional visual target acquisition algorithms. These results indicate the method's superior responsiveness and stability under natural head movement conditions. The research provides an important technical foundation for achieving natural, efficient human–computer interaction in complex environments. [ABSTRACT FROM AUTHOR]
Copyright of International Journal of Pattern Recognition & Artificial Intelligence is the property of World Scientific Publishing Company and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)*