Search Results for author:
Found 5 papers, 3 papers with code
ViPE: Visualise Pretty-much Everything
Figurative and non-literal expressions are profoundly integrated in human communication.
Visual Grounding of Inter-lingual Word-Embeddings
We obtained visually grounded vector representations for these languages and studied whether visual grounding on one or multiple languages improved the performance of embeddings on word similarity and categorization benchmarks.
How direct is the link between words and images?
To what extent does this setup rely on visual information from images?
Language with Vision: a Study on Grounded Word and Sentence Embeddings
Our model effectively balances the interplay between language and vision by aligning textual embeddings with visual information while simultaneously preserving the distributional statistics that characterize word usage in text corpora.
Learning Zero-Shot Multifaceted Visually Grounded Word Embeddings via Multi-Task Training
The general approach is to embed both textual and visual information into a common space -the grounded space-confined by an explicit relationship between both modalities.
