Image-to-video person re-identification using three-dimensional semantic appearance alignment and cross-modal interactive learning.
Wei ShiHong LiuMengyuan LiuPublished in: Pattern Recognit. (2022)
Keyphrases
- person re identification
- interactive learning
- cross modal
- three dimensional
- visual data
- input image
- image features
- image retrieval
- video surveillance
- image data
- multi modal
- semantic concepts
- visual similarity
- image classification
- image representation
- image regions
- multiscale
- image content
- visual concepts
- multimedia
- visual recognition
- test images
- metric learning
- video sequences
- spatial information
- semantic similarity
- image collections
- space time
- video data
- low level
- feature space
- similarity measure
- feature selection