Text-to-3D-Scene Generation using Semantic Parsing and Spatial Knowledge with Rule Based System

Abstract

Scene Generation plays an important role in digital media to represent a news or specific domain to the viewers. It is not easy to produce a scene from a text. Text may not complete in digital media. People are not conscious about the news until it is not visualized to them. Text to 3d Scene generation is a process where people need not read news. The 3D Scene will represent the situation. It will help people to conscious about their life. In this paper, we introduce a rule-based Framework where Scene generated from text using semantic parsing and spatial knowledge. Semantic parsing has identified the templates, objects, and constraints and spatial knowledge have built the relation between object and template. Our rule based Framework has identified the uncountable noun and some Spatial relation to generating 3d scene.