Constructing Multi-Modal Dialogue Dataset by Replacing Text with Semantically Relevant Images.
Nyoungwoo LeeSuwon ShinJaegul ChooHo-Jin ChoiSung-Hyon MyaengPublished in: ACL/IJCNLP (2) (2021)
Keyphrases
- multi modal
- relevant images
- video search
- image search
- multiple modalities
- image database
- semantic information
- image annotation
- multi modality
- image content
- keywords
- natural language
- information retrieval
- web images
- relevance feedback
- semantic concepts
- text mining
- text documents
- image registration
- image classification
- image retrieval
- text summarization
- high dimensional
- feature set