Captions Are Worth a Thousand Words: Enhancing Product Retrieval with Pretrained Image-to-Text Models.
Jason TangGarrin McGoldrickMarie Al-GhosseinChing-Wei ChenPublished in: CoRR (2024)
Keyphrases
- image retrieval
- textual descriptions
- text queries
- information retrieval
- image description
- web images
- image data
- input image
- visual features
- image content
- multiscale
- image classification
- textual and visual information
- image segmentation
- image database
- image representation
- image features
- news video
- semantic content
- image collections
- probabilistic model
- text retrieval
- visual content
- multimedia documents
- handwritten documents
- scanned documents
- auto annotation
- stop words
- retrieval systems
- image regions
- low level
- keywords
- visual similarity
- text documents
- test collection
- co occurrence
- relevance feedback