Seeing with Humans: Gaze-Assisted Neural Image Captioning.
Yusuke SuganoAndreas BullingPublished in: CoRR (2016)
Keyphrases
- input image
- image data
- neural network
- human observers
- multiscale
- image retrieval
- image content
- single image
- image segmentation
- image collections
- image representation
- image regions
- image features
- visual search
- image pixels
- image analysis
- template matching
- image classification
- network architecture
- region of interest
- eye tracking
- spatial information
- keypoints
- segmentation method
- edge detection
- segmentation algorithm
- vision system
- similarity measure
- completely automated