High-density Image Object Counting Network Based on Spatial Context and Channel Attention: Take crowds and vehicles in a statistically dense scene as an example.
Yu WangMingsheng LiuHu YiYunchi XieZhengyu TanCunyu JiPublished in: EITCE (2022)
Keyphrases
- spatial context
- image regions
- high density
- multiple objects
- single image
- temporal context
- spatial relationships
- input image
- spatial and temporal
- complex scenes
- contextual information
- spatial relations
- image content
- close proximity
- image features
- target object
- image data
- spatial information
- image segments
- low density
- multiple images
- ground plane
- multiscale
- object models
- image representation
- image classification
- consecutive frames
- low level
- appearance information
- moving objects
- lighting conditions
- keypoints
- d objects
- cast shadows
- magnetic tape
- gaussian mixture model
- space time
- feature vectors
- image retrieval
- database systems