Constructing Multi-Modal Dialogue Dataset by Replacing Text with Semantically Relevant Images.
Nyoungwoo LeeSuwon ShinJaegul ChooHo-Jin ChoiSung-Hyun MyaengPublished in: CoRR (2021)
Keyphrases
- multi modal
- relevant images
- video search
- image search
- semantic information
- multiple modalities
- image annotation
- image content
- natural language
- multi modality
- image database
- high dimensional
- text mining
- relevance feedback
- semantic concepts
- cross modal
- information retrieval
- image processing
- text documents
- keywords
- multimedia
- text summarization
- computer vision
- machine learning