Yu-Siang Wang, Chenxi Liu, Xiaohui Zeng, Alan Yuille
TL;DR本文介绍了一种通过神经依赖解析器直接从文本描述中解析出场景图来实现端到端训练的方法,并在评估数据集上比以往方法高出 5%的 F - 分数相似度来生成了我们学习的神经依赖解析器产生的场景图。
Abstract
In this paper, we study the problem of parsing structured knowledge graphs
from textual descriptions. In particular, we consider the scene graph
representation that considers objects together with their attributes and
relations: this representation has been proved useful across a varie