cs.CVNov 5, 2026

STAG-Net: Structure-Aware Attention over Landmark Groups for Lightweight Head Pose Estimation

Authors: Asiri Gawesha Lindamulage, Sanka Mohottala, Dharshana Kasthurirathna, Nuwan Kodagoda

Organizations: Faculty of Engineering, Sri Lanka Institute of Information Technology, Sri Lanka · Faculty of Engineering, University of Sri Jayewardenepura, Sri Lanka · University of Colombo School of Computing, Colombo 07, Sri Lanka · Faculty of Computing, Sri Lanka Institute of Information Technology, Sri Lanka

Abstract

Head pose estimation is important for human computer interaction and driver monitoring, but many image based methods remain too costly for resource constrained deployment. We propose STAG-Net, a structure-aware attention over landmark groups for lightweight head Pose estimation that estimates head pose from facial landmarks using movement aware selection, spatial clustering, compact DeepSet encoders, self attention, and a global geometry descriptor. We develop variants using FaceMesh and Face Alignment Network landmarks and evaluate them under cross dataset and subject independent protocols on 300W-LP, AFLW2000, and BIWI. STAG-Net achieves the lowest mean absolute error among all compared landmark based methods in every evaluated setting. On BIWI frames with valid detections, the FAN variant also records the lowest overall error among all compared methods, reaching 3.51∘3.51^\circ under cross dataset evaluation and 2.39∘2.39^\circ under subject independent evaluation. The regressors use only 0.012 to 0.043 million parameters and 24.1 to 83.6 thousand FLOPs, about three orders of magnitude fewer FLOPs than FSA-Net. Despite including landmark extraction and regression, the FaceMesh pipeline reaches 322.9 frames per second on a Tesla T4 GPU, nearly twice the 164.5 frames per second of the fastest measured image based baseline, and 111.0 frames per second on CPU. After normalization by dataset size, STAG-Net processes training samples seven to eight times faster than FSA-Net and more than one hundred times faster than HopeNet and TokenHPE, while complete training finishes within 28 minutes. These results show that STAG-Net combines leading landmark based accuracy with substantially lower training and inference cost.

Figures & tables

Appendix figures & tables4 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. VGGT-HPE: Reframing Head Pose Estimation as Relative Pose Prediction

    Apr 11, 2026Vasiliki Vasileiou, Panagiotis P. Filntisis, Petros Maragos +1Relative Pose EstimationObject Pose Estimation

  2. Bengal-HP_RU: A Dataset of Bengal People For Head Pose Estimation

    Jun 23, 2026Md. Ahanaf Arif Khan, Md. Tawhidur Rahman, Sangeeta Biswas +4

  3. SpikePose: Lightweight Video Human Pose Estimation

    Nov 4, 2026Yunhao Huang, Kaushik Bhargav Sivangi, Tianyi Yu +1Human Pose EstimationSpiking Neural Networks