A Multimodal Target-Source Classifier with Attention Branches to Understand Ambiguous Instructions for Fetching Daily Objects.
Aly MagassoubaKomei SugiuraHisashi KawaiPublished in: CoRR (2019)
Keyphrases
- focus of attention
- feature values
- target object
- object model
- multi modal
- machine learning
- d objects
- training samples
- cluttered background
- classification rate
- classification scheme
- object segmentation
- data objects
- moving objects
- training data
- decision trees
- learning algorithm
- data sets
- classification method
- visual attention
- feature set
- object classes
- training set
- support vector
- feature selection
- search engine
- target images