Representing word image using visual word embeddings and RNN for keyword spotting on historical document images.
Hongxi WeiHui ZhangGuanglai GaoPublished in: ICME (2017)
Keyphrases
- document images
- keyword spotting
- printed documents
- visual words
- bag of words
- handwritten documents
- image classification
- document analysis
- document image analysis
- co occurrence
- image representation
- optical character recognition
- keypoints
- document processing
- image retrieval
- visual content
- image features
- character recognition
- text lines
- hidden markov models
- vector space
- language independent
- speech recognition
- information retrieval
- multiscale
- feature space
- gaussian mixture model
- n gram
- text classification
- feature selection