首页> 外文会议>International natural language generation conference >Towards Generating Colour Terms for Referents in Photographs: Prefer the Expected or the Unexpected?
【24h】

Towards Generating Colour Terms for Referents in Photographs: Prefer the Expected or the Unexpected?

机译:争取为照片中的对象生成颜色术语:更喜欢预期还是意外?

获取原文

摘要

Colour terms have been a prime phenomenon for studying language grounding, though previous work focussed mostly on descriptions of simple objects or colour swatches. This paper investigates whether colour terms can be learned from more realistic and potentially noisy visual inputs, using a corpus of referring expressions to objects represented as regions in real-world images. We obtain promising results from combining a classifier that grounds colour terms in visual input with a recalibra-tion model that adjusts probability distributions over colour terms according to contextual and object-specific preferences.
机译:颜色术语一直是研究语言基础的主要现象,尽管先前的工作主要集中在对简单对象或颜色样本的描述上。本文研究了通过使用将表达式引用到真实世界图像中表示为区域的对象的语料库,是否可以从更逼真的,可能更嘈杂的视觉输入中学习颜色术语。通过将基于视觉输入中的颜色术语的分类器与根据上下文和特定于对象的偏好调整颜色术语的概率分布的重新校准模型相结合,我们获得了令人鼓舞的结果。

著录项

相似文献

  • 外文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号