Generate Image Descriptions based on Deep RNN and Memory Cells for Images Features.
Shijian TangSong HanPublished in: CoRR (2016)
Keyphrases
- image features
- original images
- test images
- image description
- input image
- image data
- image regions
- image collections
- sample images
- image set
- image pixels
- extracting features
- image retrieval
- grey levels
- image statistics
- keypoints
- image similarity
- image classification
- visual appearance
- grey level
- filtered images
- image analysis
- color histogram
- global features
- extracted features
- image matching
- image content
- feature detectors
- edge detection
- feature vectors
- relevant images
- region of interest
- image representation
- pixel intensities
- feature points
- image database
- feature matching
- pixel values
- target object
- matching process
- color features
- spatial information
- segmentation method
- microscope images
- textural features
- affine moment invariants
- texture and shape features
- invariant features
- nearest neighbor
- intensity images
- sift descriptors
- spatial relationships
- texture features
- lighting conditions
- labeled images
- natural images
- single image
- histological images
- microscopy images
- segmentation algorithm
- human observers
- gabor filters
- feature descriptors
- contrast enhancement
- recurrent neural networks
- visual features
- enhanced image
- geometric constraints
- scale space
- image descriptors
- feature space
- object recognition
- web images
- human retina
- feature extraction
- image segmentation