Improving Person Re-identification by Background Subtraction Using Two-Stream Convolutional Networks.
Mahmoud GhorbelSourour AmmarYousri KessentiniMohamed JmaielPublished in: ICIAR (1) (2019)
Keyphrases
- background subtraction
- person re identification
- video surveillance
- sparse coding
- surveillance system
- moving objects
- object detection
- image sequences
- foreground objects
- data streams
- gaussian mixture model
- multiple cameras
- camera network
- appearance model
- dynamic scenes
- post processing
- linear combination
- generative model
- event detection
- computer vision
- multiscale
- sensor networks
- unsupervised learning
- sparse representation
- image representation