Scene Graph as Pivoting: Inference-time Image-free Unsupervised Multimodal Machine Translation with Visual Scene Hallucination.
Hao FeiQian LiuMeishan ZhangMin ZhangTat-Seng ChuaPublished in: CoRR (2023)
Keyphrases
- visual scene
- machine translation
- complex scenes
- input image
- single image
- multiscale
- image features
- image collections
- image data
- vision system
- natural images
- object recognition
- spatial relations
- visual information
- image regions
- information extraction
- visual attention
- cross lingual
- image content
- statistical machine translation
- image representation
- natural language processing
- video sequences
- natural language
- cross language information retrieval
- multiple images
- multiple objects
- artificial intelligence
- image classification
- target language
- image set
- high resolution
- image processing
- computer vision
- natural scenes
- image segmentation
- image sequences
- image retrieval
- feature vectors