Fusing object detection and region appearance for image-text alignment.
Luca Del PeroPhilip LeeJames MagahernEmily HartleyKobus BarnardPing WangAtul KanaujiaNiels HaeringPublished in: ACM Multimedia (2011)
Keyphrases
- input image
- object detection
- image regions
- image features
- region of interest
- multiscale
- grey level
- scene understanding
- image data
- target object
- single image
- fusion method
- face detection
- image pixels
- image segmentation
- image alignment
- web images
- high resolution
- bounding box
- outdoor scenes
- static images
- image content
- homogeneous regions
- appearance cues
- adjacent regions
- segmentation algorithm
- image retrieval
- segmented images
- region features
- pixel wise
- text regions
- image classification
- image representation
- edge map
- background subtraction
- face images
- test images
- edge detection
- region segmentation
- boundary information
- segmentation method
- camera viewpoint
- text graphics