LISA: Language-guided Interference-aware Spatial-Frequency Attention for Driver Gaze Estimation

Ma, Jun; Yang, Zhenye; Zhou, Ruichen; Zhang, Pei; Li, Huan; Chen, Jinpeng

doi:10.24963/ijcai.2026/325

LISA: Language-guided Interference-aware Spatial-Frequency Attention for Driver Gaze Estimation

Jun Ma, Zhenye Yang, Ruichen Zhou, Pei Zhang, Huan Li, Jinpeng Chen

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 2924-2932.

https://doi.org/10.24963/ijcai.2026/325

Keywords

Data Mining: Applications
Machine Learning: Multi-modal learning

Abstract

Driver gaze estimation serves as a fundamental metric for evaluating driver attentiveness in modern monitoring systems. Beyond being vulnerable to sudden lighting changes and sensor noise, spatial-domain models struggle to disentangle authentic gaze cues from irrelevant visual attributes. In this paper, we propose LISA, a Language-guided Interference-aware Spatial-Frequency Attention framework that combines frequency-domain priors with vision-language knowledge. Observing that the amplitude spectrum remains relatively stable even under spatial perturbations, we design a dual-domain fusion mechanism. It integrates stable low-frequency semantics into high-frequency details, employing spatial attention to precisely target ocular regions. To reduce semantic ambiguity, we also introduce a training-time disentanglement strategy. Using a frozen CLIP encoder and orthogonal regularization, we explicitly separate gaze features from appearance interference. Experiments on two benchmarks show that LISA achieves state-of-the-art performance, with significantly improved robustness against occlusions and lighting variations. The code repository is available at https://github.com/Mason-bupt/LISA.

Jun Ma, Zhenye Yang, Ruichen Zhou, Pei Zhang, Huan Li, and Jinpeng Chen. 2026. LISA: Language-guided Interference-aware Spatial-Frequency Attention for Driver Gaze Estimation. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 2924–2932. https://doi.org/10.24963/ijcai.2026/325

@inproceedings{ijcai2026p325,
  title     = {LISA: Language-guided Interference-aware Spatial-Frequency Attention for Driver Gaze Estimation},
  author    = {Ma, Jun and Yang, Zhenye and Zhou, Ruichen and Zhang, Pei and Li, Huan and Chen, Jinpeng},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {2924--2932},
  year      = {2026},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2026/325},
  url       = {https://doi.org/10.24963/ijcai.2026/325},
}

TY  - CPAPER
TI  - LISA: Language-guided Interference-aware Spatial-Frequency Attention for Driver Gaze Estimation
AU  - Ma, Jun
AU  - Yang, Zhenye
AU  - Zhou, Ruichen
AU  - Zhang, Pei
AU  - Li, Huan
AU  - Chen, Jinpeng
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 2924
EP  - 2932
DO  - 10.24963/ijcai.2026/325
UR  - https://doi.org/10.24963/ijcai.2026/325
ER  -