Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models.
Bryan A. PlummerLiwei WangChris M. CervantesJuan C. CaicedoJulia HockenmaierSvetlana LazebnikPublished in: CoRR (2015)
Keyphrases
- input image
- image features
- image retrieval
- image collections
- region of interest
- image regions
- single image
- high resolution
- image data
- multiscale
- homogeneous regions
- grey level
- image content
- image pixels
- image structure
- adjacent regions
- region segmentation
- bayesian framework
- image representation
- gradient information
- segmentation method
- feature matching
- image segmentation
- segmented images
- probabilistic model
- parametric models
- edge detection
- image set
- noun phrases
- image classification
- natural language
- salient regions
- linguistic features
- boundary information
- edge map
- feature points
- object recognition
- social media
- low level
- image matching