PVT-Crowd: Bridging Multi-scale Features from Pyramid Vision Transformer for Weakly-Supervised Crowd Counting.
Zhanqiang HuoKunwei ZhangFen LuoYingxu QiaoPublished in: PRCV (9) (2023)
Keyphrases
- multiscale
- weakly supervised
- superpixels
- feature extraction
- image representation
- scale space
- feature set
- computer vision
- low level
- feature space
- image features
- question answering
- topic models
- object class
- relation extraction
- information retrieval
- semi supervised
- co occurrence
- higher order
- spatial information
- image segmentation