计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (6A): 250300095-7.doi: 10.11896/jsjkx.250300095

• 人工智能 • 上一篇    下一篇

基于TCN-AttnRNN模型的眼动输入动态调整技术

陈迪, 殷继彬   

  1. 昆明理工大学信息工程与自动化学院 昆明 650500
  • 出版日期:2026-06-16 发布日期:2026-06-12
  • 通讯作者: 殷继彬(yjblovelh@aliyun.com)
  • 作者简介:(3458959449@qq.com)

Dynamic Adjustment Technology of Eye Movement Input Based on TCN-AttnRNN Model

CHEN Di, YIN Jibin   

  1. Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China
  • Published:2026-06-16 Online:2026-06-12
  • About author:CHEN Di,born in 1998,postgraduate.His main research interests include deep learning and human-computer interaction.
    YIN Jibin,born in 1976,Ph.D,associate professor.His main research interests include human-computer interaction and artificial intelligence.

PDF (PC)

37

摘要： 文中提出了一种根据字符预测结果动态调整键驻留时间的眼动输入技术。在本研究中,设计了一种字符级语言模型,命名为TCN-AttnRNN。在该模型中,TCN负责提取序列的全局空间特征和长期依赖关系,RNN增强时间序列的长期记忆性能,而多头自注意力机制则优化网络,通过概率分配权重来增强关键特征的作用。实验结果表明,在PTB数据集和DailyDialog数据集上,TCN-AttnRNN模型的BPC值分别为1.26和1.22,优于当前主流的TCN,LSTM和Transformer模型。在TCN-AttnRNN模型的基础上,设计了一种眼动输入动态调整技术。通过使用TCN-AttnRNN模型进行字符预测,该技术根据用户下一次选择按键的概率来调整其驻留时间。实验结果证实了该技术的有效性,相较于传统的固定驻留时间方法,该技术使得用户的输入速度提高了22.31%,同时将校正错误率降低了19.57%。

关键词: 人机交互, 深度学习, 语言建模, 字符级语言模型, 眼动输入

Abstract: This paper presents an eye movement input technology that dynamically adjusts key dwell time based on character prediction results.In this study,a character-level language model named TCN-AttnRNN is designed.In this model,TCN is responsible for extracting global spatial features and long-term dependencies of sequences,RNN enhances the long-term memory performance of time series,and the multi-head self-attention mechanism optimizes the network by allocating weights through probability distribution to enhance the role of key features.Experimental results show that the BPC values of the TCN-AttnRNN model on the PTB and DailyDialog datasets are 1.26 and 1.22 respectively,which are superior to the current mainstream TCN,LSTM,and Transformer models.Based on the TCN-AttnRNN model,a dynamic adjustment technology for eye movement input is designed.By using the TCN-AttnRNN model for character prediction,this technology adjusts the dwell time of keys according to the pro-bability of users' next key selection.Experimental results confirm the effectiveness of this technology,compared with the traditional fixed dwell time method,it increases users' input speed by 22.31% and reduces the correction error rate by 19.57%.

Key words: Human-computer interaction, Deep learning, Language modeling, Character-level language model, Eye movement input

中图分类号: 

  • TP391
[1] DUCHOWSKI A T.Gaze-based interaction:A 30 year retro-spective[J].Computers & Graphics,2018,73:59-69.
[2] ARMSTRONG T,OLATUNJI B O.Eye tracking of attention in the affective disorders:A meta-analytic review and synthesis[J].Clinical Psychology Review,2012,32(8):704-723.
[3] CAO X.Eye tracking in human-computer interaction recognition[C]//2023 IEEE International Conference on Sensors,Electronics and Computer Engineering(ICSECE).IEEE,2023:203-207.
[4] JIANG Z,XU F F,ARAKI J,et al.How can we know what language models know?[J].Transactions of the Association for Computational Linguistics,2020,8:423-438.
[5] PASCANU R,GULCEHRE C,CHO K,et al.How to construct deep recurrent neural networks[J].arXiv:1312.6026,2013.
[6] SCHMIDHUBER J,HOCHREITER S.Long short-term memory[J].Neural Computation,1997,9(8):1735-1780.
[7] KRAUSE B,LU L,MURRAY I,et al.Multiplicative LSTM for sequence modelling[J].arXiv:1609.07959,2016.
[8] MUJIKA A,MEIER F,STEGER A.Fast-slow recurrent neural networks[J].Advances in Neural Information Processing Systems,2018,30:5916-5925.
[9] VASWANI A,SHAZEER N,PARMAR N,et al.Attention isall you need[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems.Curran Associates Inc.,2017:6000-6010.
[10] DAI Z,YANG Z,YANG Y,et al.Transformer-xl:Attentive language models beyond a fixed-length context[J].arXiv:1901.02860,2019.
[11] ŠPAKOV O,MINIOTAS D.On-line adjustment of dwell time for target selection by gaze[C]//Proceedings of the third Nordic conference on Human-computer interaction.2004:203-206.
[12] MAJARANTA P,AULA A,RÄIHÄ K J.Effects of feedbackon eye typing with a short dwell time[C]//Proceedings of the 2004 Symposium on Eye Tracking Research & Applications.2004:139-146.
[13] KRISTENSSON P O,VERTANEN K.The potential of dwell-free eye-typing for fast assistive gaze communication[C]//Proceedings of the Symposium on Eye Tracking Research and Applications.2012:241-244.
[14] PEDROSA D,PIMENTEL M D G,WRIGHT A,et al.Fil-teryedping:Design challenges and user performance of dwell-free eye typing[J].ACM Transactions on Accessible Computing(TACCESS),2015,6(1):1-37.
[15] LIU Y,ZHANG C,LEE C,et al.Gazetry:Swipe text typing using gaze[C]//Proceedings of the Annual Meeting of the Australian Special Interest Group for Computer-Human Interaction.2015:192-196.
[16] KURAUCHI A,FENG W,JOSHI A,et al.EyeSwipe:Dwell-free text entry using gaze paths[C]//Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems.2016:1952-1956.
[17] WARD D J,BLACKWELL A F,MACKAY D J C.Dasher－a data entry interface using continuous gestures and language models[C]//Proceedings of the 13th annual ACM symposium on User interface software and technology.2000:129-137.
[18] HANSEN D W,SKOVSGAARD H H T,HANSEN J P,et al.Noise tolerant selection by gaze-controlled pan and zoom in 3D[C]//Proceedings of the 2008 Symposium on Eye Tracking Research & Applications.2008:205-212.
[19] HUCKAUF A,URBINA M H.Gazing with pEYEs:towards a universal input for various applications[C]//Proceedings of the 2008 symposium on Eye tracking research & applications.2008:51-54.
[20] LAURIOLA I,LAVELLI A,AIOLLI F.An introduction todeep learning in natural language processing:Models,techniques,and tools[J].Neurocomputing,2022,470:443-456.
[21] LI Y,SU H,SHEN X,et al.Dailydialog:A manually labelled multi-turn dialogue dataset[J].arXiv:1710.03957,2017.
[22] TRAN D,VAFA K,AGRAWAL K K,et al.Discrete flows:invertible generative models of discrete data[C]//Proceedings of the 33rd International Conference on Neural Information Processing Systems.2019:14719-14728.
[23] TURKOGLU M O,D′ARONCO S,WEGNER J D,et al.Gating revisited:Deep multi-layer RNNs that can be trained[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2021,44(8):4081-4092.
[24] MOTT M E,WILLIAMS S,WOBBROCK J O,et al.Improving dwell-based gaze typing with dynamic,cascading dwell times[C]//Proceedings of the 2017 CHI Conference on Human Factors in Computing Systems.2017:2558-2570.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发