Crossmodal Sound Retrieval Based on Specific Target Co-Occurrence Denoted with Weak Labels.
Masahiro YasudaYasunori OhishiYuma KoizumiNoboru HaradaPublished in: INTERSPEECH (2020)
Keyphrases
- co occurrence
- term co occurrence
- query terms
- image database
- named entities
- information retrieval
- pairwise
- image retrieval
- wordnet
- latent semantic analysis
- domain specific
- term weights
- semantic relations
- semantic similarity
- text retrieval
- artificial intelligence
- machine learning
- multiple types
- visual words
- topic models
- information retrieval systems
- relevance feedback
- retrieval model
- document retrieval
- occurrence frequency