Generation of Relative Referring Expressions based on Perceptual Grouping
Generation of Relative Referring Expressions based on Perceptual Grouping
复制标题
基于感知分组的相对指代表达的生成
DOI：
10.3115/1220355.1220451
复制
发表时间：
2004
期刊：
影响因子：
--
通讯作者：
T. Tokunaga
中科院分区：
文献类型：
--
作者：
Kotaro Funakoshi;Satoru Watanabe;Naoko Kuriyama;T. Tokunaga
过去生成指称表达式的工作主要是利用对象的属性和对象之间的二进制关系。然而，当对象之间没有明显的属性时，这种方法不能很好地工作。为了克服这一局限性，本文提出了一种利用对象的感知群和它们之间的n元关系的方法。关键是识别人类自然识别的对象组。我们对42名受试者进行了心理实验，收集了这种情景下的指称表情，并根据实验结果构建了一个生成算法。对另外23名受试者的评估表明，该方法能够有效地生成正确的指称表达。
Past work of generating referring expressions mainly utilized attributes of objects and binary relations between objects. However, such an approach does not work well when there is no distinctive attribute among objects. To overcome this limitation, this paper proposes a method utilizing the perceptual groups of objects and n-ary relations among them. The key is to identify groups of objects that are naturally recognized by humans. We conducted psychological experiments with 42 subjects to collect referring expressions in such situations, and built a generation algorithm based on the results. The evaluation using another 23 subjects showed that the proposed method could effectively generate proper referring expressions.