VISA: An Ambiguous Subtitles Dataset for Visual Scene-Aware Machine Translation.
Yihang LiShuichiro ShimizuWeiqi GuChenhui ChuSadao KurohashiPublished in: CoRR (2022)
Keyphrases
- machine translation
- visual scene
- language independent
- natural language processing
- vision system
- cross language information retrieval
- object recognition
- information extraction
- cross lingual
- complex scenes
- visual attention
- target language
- natural language
- language resources
- word sense disambiguation
- visual information
- word alignment
- chinese english
- machine translation system
- natural images
- natural scenes
- statistical machine translation
- artificial intelligence
- text classification
- real time
- spatial relations
- eye movements
- feature selection