Tackling background ambiguities in multi-class few-shot point cloud semantic segmentation.
Lvlong LaiJian ChenChi ZhangZehong ZhangGuosheng LinQingyao WuPublished in: Knowl. Based Syst. (2022)
Keyphrases
- multi class
- point cloud
- semantic segmentation
- object detection
- pascal voc
- conditional random fields
- superpixels
- weakly supervised
- structure from motion
- scene classification
- object categories
- object class
- support vector machine
- pairwise
- object classes
- bounding box
- bag of words
- video sequences
- visual features
- feature selection
- data sets
- input image
- viewpoint
- long range
- image classification