Augmenters at SemEval-2023 Task 1: Enhancing CLIP in Handling Compositionality and Ambiguity for Zero-Shot Visual WSD through Prompt Augmentation and Text-To-Image Diffusion.
Jie S. LiYow-Ting ShiueYong-Siang ShihJonas GeipingPublished in: CoRR (2023)
Keyphrases
- web images
- word sense disambiguation
- image data
- image features
- input image
- multiscale
- low level
- visual appearance
- image classification
- test images
- image content
- web image search
- visually similar
- visual perception
- visual data
- image representation
- image segmentation
- single image
- high resolution
- visual cues
- image processing
- textual descriptions
- image collections
- edge detection
- image retrieval
- diffusion process
- similarity measure
- visual concepts
- spatial relations
- visual features
- visual information
- information retrieval