Weakly-supervised video re-localization with multiscale attention model

Keyphrases

• Attention Model 100%
• Unsupervised Learning 100%
• Reference Video 100%
• Video Re-localization 100%
• Multi-scale Attention 100%
• Fully-supervised 28%
• Feature Extraction 14%
• Public Dataset 14%
• Attention Mechanism 14%
• Subsequence 14%
• Delocalization 14%
• CNN Model 14%
• Art Performance 14%
• Co-attention 14%
• C3D Network 14%
• CNN-based Model 14%
• Multi-temporal Features 14%

Computer Science

• Convolutional Neural Network 100%
• Temporal Feature 50%
• Attention (Machine Learning) 50%
• Art Performance 50%
• Feature Extraction 50%