ACL RD-TEC 1.0 Summarization of W03-0604
Paper Title:
TOWARDS A FRAMEWORK FOR LEARNING STRUCTURED SHAPE MODELS FROM TEXT-ANNOTATED IMAGES
TOWARDS A FRAMEWORK FOR LEARNING STRUCTURED SHAPE MODELS FROM TEXT-ANNOTATED IMAGES
Authors: Sven Wachsmuth and Suzanne Stevenson and Sven Dickinson
Primarily assigned technology terms:
- 2-d shape recognition
- algorithm
- approximation
- bootstrapping
- chunker
- computational linguistics
- computer vision
- crawler
- database
- distance function
- generic object recognition
- grouping
- image segmentation
- iterative bootstrapping
- learning
- matching
- matching algorithm
- mean-shift region segmentation
- object recognition
- probabilistic translation
- processing
- ranking
- recognition
- region segmentation
- scene generation
- scoring
- segmentation
- segmentation algorithm
- shape recognition
- structuring
- tagger
- vector quantization
- world wide web
Other assigned terms:
- ambiguity
- approach
- bilingual text
- case
- class name
- clusters
- compounds
- computational complexity
- conditional probability
- contour
- data set
- data sparseness
- distribution
- estimation
- experimental results
- fact
- feature
- feature space
- feature vector
- finite set
- french
- french word
- generation
- implementation
- joint distribution
- knowledge
- labeling
- language model
- lexicon
- linguistic
- linguistics
- local maximum
- many-to-many mapping
- mapping
- mappings
- maps
- measures
- mechanisms
- method
- mutual information
- names
- noun phrases
- nouns
- object class
- part-ofspeech
- part-whole relation
- priori
- probabilities
- probability
- queries
- ranking scheme
- relation
- representations
- semantic
- semantic hierarchy
- semantic information
- semantic structure
- shock graph
- structured object model
- structured text
- symbol
- symbols
- syntactic structure
- terms
- text
- training
- transformation
- translation model
- translation models
- translation table
- visual information
- vocabulary
- word
- word sequences
- words
