A comparison of dense region detectors for image search and fine-grained classification.
Ahmet IscenGiorgos ToliasPhilippe Henri GosselinHervé JégouPublished in: CoRR (2014)
Keyphrases
- fine grained
- image search
- image classification
- coarse grained
- visual features
- image collections
- feature extraction
- image retrieval
- image annotation
- video search
- decision trees
- classification accuracy
- machine learning
- feature space
- feature selection
- object detection
- relevance feedback
- feature vectors
- access control
- web images
- higher order
- web image search
- text classification
- natural language processing
- computer vision
- image regions
- feature set
- image features
- multiscale
- information retrieval