中文摘要：

针对基于视觉的智能电视手势交互中用户认知负荷和操作负荷较重的问题,提出融合显式交互和隐式交互的手势交互算法.通过对基于视觉的智能电视手势交互场景分析,首先建立了基于用户行为和智能电视状态的多层次上下文模型,实现上下文的数据融合与特征提取;其次设计并实现了CDL-DFCM推理模型和显隐信息融合的隐式交互模型,识别交互情景事件并感知用户意图;最后实现用户与智能电视的隐式交互.实验结果表明,与现有算法相比,该算法在操作准确率、时间开销和手势移动距离等方面得到了明显改善,并有效地提升了用户体验.

英文摘要：

Since user’s cognitive load and operational load is heavy in the visual gesture interaction for smart TV, we propose an implicit interaction algorithm which is based on explicit and implicit interactive context.Firstly, by analyzing the smart TV’s gesture interactive scene based on the vision, we establish a multi-levelcontext model based on user’s behavior and smart TV’s status, and implement the data’s fusion and feature’sextraction of interactive context. Then, we design and implement the inference model of CDL-DFCM andimplicit interaction model which is based on the data fusion of the explicit and implicit context. The proposedmethod can identify the situational event of interaction and perceive the user’s intention. Finally, the implicit interaction between user and smart TV is realized. Compared with the existing algorithms, experimental results show that this algorithm can improve the accuracy of operations, and reduce the time spending and the gesture’s movement distance. User experience has also been promoted.