Interactive Natural Language Grounding via Referring Expression Comprehension and Scene Graph Parsing.
Jinpeng MiJianzhi LyuSong TangQingdu LiJianwei ZhangPublished in: Frontiers Neurorobotics (2020)
Keyphrases
- natural language
- natural language processing
- d scene
- natural language interface
- graph grammars
- semantic representation
- question answering
- knowledge representation
- three dimensional
- video sequences
- linguistic analysis
- directed graph
- graph model
- live video
- graph structure
- language processing
- language understanding
- weighted graph
- multiple images
- machine learning
- graph theory
- natural language sentences
- syntactic structures
- graph representation
- computer graphics
- dialogue system
- information extraction
- highly ambiguous
- multiple objects
- graph matching
- outdoor scenes
- image sequences
- directed acyclic graph
- natural language parsing
- semantic interpretation
- scene understanding
- spanning tree
- graph databases
- graph mining
- bipartite graph
- connected components
- image segmentation