Embodied Visual Active Learning for Semantic Segmentation.
David NilssonAleksis PirinenErik GärtnerCristian SminchisescuPublished in: CoRR (2020)
Keyphrases
- semantic segmentation
- active learning
- street scenes
- conditional random fields
- superpixels
- scene classification
- weakly supervised
- object categories
- label transfer
- machine learning
- pascal voc
- learning algorithm
- semi supervised
- training examples
- data sets
- object class
- object recognition
- learning process
- object classes
- visual features
- supervised learning
- object detection
- input image
- viewpoint
- segmentation algorithm
- pixel level
- support vector
- multiscale
- feature selection
- computer vision