Combining image and text features: a hybrid approach to mobile book spine recognition.
Sam S. TsaiDavid M. ChenHuizhong ChenCheng-Hsin HsuKyu-Han KimJatinder Pal SinghBernd GirodPublished in: ACM Multimedia (2011)
Keyphrases
- image features
- extracting features
- feature extractor
- static images
- low level
- image data
- visual object recognition
- spatial information
- input image
- feature extraction
- handwritten words
- object recognition
- image content
- object models
- multiscale
- recognizing objects
- single image
- test images
- image classification
- extracted features
- image matching
- invariant features
- keypoints
- connected component analysis
- recognition rate
- image retrieval
- document analysis
- object representation
- feature values
- image segmentation
- image set
- image regions
- feature vectors
- highly discriminative
- action recognition
- digital imaging
- image representation
- partial occlusion
- high resolution
- printed documents
- text lines
- image collections
- web images
- handwriting recognition
- mobile phone
- image sequences
- computer vision
- occluded objects
- camera phones
- text messaging
- texture and shape features