Year: 2025
Type: article
Abstract: This paper proposes a lightweight gaze-tracking architecture named GhostBlock-Augmented Look to Coordinate Space (L2CS), which integrates GhostNet-based modules and depthwise separable convolution to achieve a better trade-off between model accuracy and computational efficiency. Conventional lightweight gaze-tracking models often suffer from degraded accuracy due to aggressive parameter reduction. To address this issue, we introduce GhostBlocks, a custom-designed convolutional unit that combines intrinsic feature generation with ghost feature recomposition through depthwise operations. Our method enhances the original L2CS architecture by replacing each ResNet block with GhostBlocks, thereby significantly reducing the number of parameters and floating-point operations. The experimental results on the Gaze360 dataset demonstrate that the proposed model reduces FLOPs from 16.527 × 108 to 8.610 × 108 and parameter count from 2.387 × 105 to 1.224 × 105 while maintaining comparable gaze estimation accuracy, with MAE increasing only slightly from 10.70° to 10.87°. This work highlights the potential of GhostNet-augmented designs for real-time gaze tracking on edge devices, providing a practical solution for deployment in resource-constrained environments.
Source: Electronics
Authors: Jing-Ming Guo, Yu-Sung Cheng, Yi-Chong Zeng, Zong-Yan Yang
Institution: National Taiwan University of Science and Technology
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,896
OpenAlex ID: W4413272135