I2MVFormer: Large Language Model Generated Multi-View Document Supervision for Zero-Shot Image Classification.
Muhammad Ferjad NaeemMuhammad Gul Zain Ali KhanYongqin XianMuhammad Zeshan AfzalDidier StrickerLuc Van GoolFederico TombariPublished in: CVPR (2023)
Keyphrases
- multi view
- language model
- image classification
- document retrieval
- document ranking
- ad hoc information retrieval
- information retrieval
- language modeling
- query terms
- vector space model
- single view
- document length
- word clouds
- retrieval model
- n gram
- probabilistic model
- query specific
- three dimensional
- query expansion
- mixture model
- bag of words
- language modeling framework
- d objects
- image features
- relevance model
- smoothing methods
- feature extraction
- document collections
- image representation
- semi supervised
- visual features
- test collection
- relevant documents
- web documents
- text documents
- multi label
- co training
- retrieval systems
- translation model
- information retrieval systems
- active learning
- viewpoint
- document clustering
- tf idf
- keywords
- collaborative filtering
- decision trees