Person re-identification with multiple similarity probabilities using deep metric learning for efficient smart security applications

Keyphrases

• Person Re-identification 100%
• Security Applications 100%
• Deep Metric Learning 100%
• Smart Security 100%
• Similarity Probability 100%
• Multiple Similarity 100%
• Classification Network 50%
• Stacked Autoencoder Network 50%
• Existing Algorithms 25%
• Time Complexity 25%
• Learning Methods 25%
• Metric Learning 25%
• Dynamic Environment 25%
• Selected Features 25%
• Re-identification 25%
• Pixel Level 25%
• Public Dataset 25%
• Computing Capabilities 25%
• Real Environment 25%
• Pedestrian 25%
• Learning Objectives 25%
• Similarity Relation 25%
• Single Metrics 25%
• Multiple Metrics 25%
• AdaBoost 25%
• Smart City 25%
• Surveillance Video Analysis 25%
• Encoder Classification 25%
• Softmax Regression 25%
• Intelligent Surveillance Networks 25%
• Probability Measurement 25%
• Surveillance Imaging 25%
• Identification Technology 25%
• Daily Operation 25%
• Metric Function 25%

Computer Science

• Person Re-Identification 100%
• Security Application 100%
• Autoencoder 100%
• Experimental Result 50%
• Time Complexity 50%
• Dynamic Environment 50%
• Reidentification 50%
• surveillance video 50%
• Smart City 50%
• Metric Function 50%