Weakly Supervised 3D Semantic Segmentation Using Cross-Image Consensus and Inter-Voxel Affinity Relations.
Xiaoyu ZhuJeffrey ChenXiangrui ZengJunwei LiangChengqi LiSinuo LiuSima BehpourMin XuPublished in: ICCV (2021)
Keyphrases
- semantic segmentation
- weakly supervised
- superpixels
- conditional random fields
- object class
- scene classification
- object classes
- topic models
- multiscale
- relation extraction
- object categories
- input image
- object recognition
- image content
- pascal voc
- image set
- pairwise
- bounding box
- image processing
- image features
- semi supervised
- image representation
- automatic extraction
- image retrieval
- semantic relations
- region of interest
- bag of words
- image classification