Scene Theme Extraction Based on Scene Lexicon
- 10.2991/amcce-18.2018.33How to use a DOI?
- text-to-scene conversion, scene theme, extraction of scene theme, scene lexicon.
With the advent of the era of big data, the computing efficiency of computers has become higher and higher, and the artificial intelligence technology has been booming. These technologies have continuously developed to provide technical support for the text-to-scene conversion and brought new challenges at the same time. In the text-to-scene conversion system, there is a problem about that a text-to-scene conversion system can generate rich and varied scenes if the text input is strictly limited; if the strict restrictions on input are eliminated, the system can convert a vary kinds of scene. Therefore, if we want to a text-to-scene system can generate rapidly all kinds of scene, regardless of the length and the writing format of the articles, we need to know what scene theme of this article is. When we know the scene theme of this article, we can get the essential information when scenes are creating, such as the common entities under this scene, the attributes of common entities and spatial relationships of common entities. This paper brings out a notion of scene theme and expatiates a method to extract the scene theme from the article.
- © 2018, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Fuping Yang AU - Yuke Bai PY - 2018/05 DA - 2018/05 TI - Scene Theme Extraction Based on Scene Lexicon BT - Proceedings of the 2018 3rd International Conference on Automation, Mechanical Control and Computational Engineering (AMCCE 2018) PB - Atlantis Press SP - 189 EP - 193 SN - 2352-5401 UR - https://doi.org/10.2991/amcce-18.2018.33 DO - 10.2991/amcce-18.2018.33 ID - Yang2018/05 ER -