MsVFE and V-SIAM: Attention-based multi-scale feature interaction and fusion for outdoor LiDAR semantic segmentation.
Jingru YangJin WangKaixiang HuangGuodong LuYu SunHuan YuCheng ZhangYing YangWenming ZouPublished in: Neurocomputing (2024)
Keyphrases
- semantic segmentation
- multiscale
- superpixels
- street scenes
- image fusion
- label transfer
- conditional random fields
- weakly supervised
- pascal voc
- natural images
- visual attention
- object class
- object categories
- image understanding
- pixel level
- local binary pattern
- bag of words
- image representation
- image features
- segmentation algorithm
- multiple images
- long range
- scale space
- object detection
- scene classification
- probabilistic model
- feature vectors
- viewpoint
- object recognition