One Class One Click: Quasi Scene-level Weakly Supervised Point Cloud Semantic Segmentation with Active Learning.
Puzuo WangWei YaoJie ShaoPublished in: CoRR (2022)
Keyphrases
- weakly supervised
- semantic segmentation
- point cloud
- scene classification
- active learning
- semi supervised
- superpixels
- conditional random fields
- topic models
- object recognition
- multiple images
- object class
- relation extraction
- image set
- object detectors
- input image
- three dimensional
- named entities
- supervised learning
- learning algorithm
- image representation
- video sequences
- multiscale
- visual words
- multiple objects
- training examples
- single image
- labeled data
- image classification