中文摘要：

提出一个在线双向适应的笔手势界面框架,该框架针对传统笔手势界面中静态手势识别器不能支持用户的个性化输入以及用户在笔手势界面中面临的手势记忆问题,提出了在线双向适应的策略：一方面系统能够适应用户（系统可以在线支持用户的个性化输入）;另一方面用户可以学习系统（用户可以学习系统提供的某些笔手势）.该框架包括5个部分：（1）双向适应笔手势输入解释模型;（2）双向适应笔手势输入解释流程;（3）上下文优先级定义;（4）纠错和模糊消解界面;（5）在线笔手势查询帮助系统.在该框架的指导下作者开发了一个原型系统并进行了对比实验评估.结果表明该框架在可用性上具有较大的优势.

英文摘要：

This paper presents a framework for on-line two-way adaptive pen gestures based user interface.In order to solve the problem that the traditional stationary pen gestures recognition engine can＇t support users＇ personified gestures input habits and the problem that users maybe can＇t remember some gestures when using them,the authors propose an on-line two-way adaptive strategy,on the one hand,system can adapt to different users according to their different input habits（system supports on-line personified pen gestures input habits through mediate interface）,on the other hand,users can learn the interpretation capability of the system（users can learn some system gestures through mediate interface or on-line pen gestures query and help system）.The framework consists of five parts：（1） on-line two-way adaptive pen gestures input interpretation model;（2） on-line two-way adaptive pen gestures input interpretation flowchart;（3） contextual priority definitions;（4） context-based mediate interface;（5） on-line context-based pen gestures query and help system.The authors implement a prototype under the guidance of the framework;in such prototype users can input gestures to manipulate photos.A controlled comparative quantitative experiment based on the prototype is carried out.The result of the experiment validates the advantages of the framework in usability.