Enhancing CLIP with GPT-4: Harnessing Visual Descriptions as Prompts.
Mayug ManiparambilChris VorsterDerek MolloyNoel MurphyKevin McGuinnessNoel E. O'ConnorPublished in: CoRR (2023)
Keyphrases
- visual information
- high level
- databases
- video clips
- visual features
- neural network
- low level
- visual perception
- visual analysis
- visual exploration
- object recognition
- artificial neural networks
- hidden markov models
- multi modal
- social networks
- real time
- visual search
- visual cues
- mental models
- visual representation
- visual processing