TASTNet: An end-to-end deep fingerprinting net with two-dimensional attention mechanism and spatio-temporal weighted fusion for video content authentication.
Gejian ZhaoFengyong LiHeng YaoChuan QinPublished in: J. Vis. Commun. Image Represent. (2023)
Keyphrases
- end to end
- spatio temporal
- attention mechanism
- scalable video
- spatial and temporal
- space time
- human actions
- video sequences
- visual attention
- video frames
- video data
- moving objects
- video content
- video streams
- congestion control
- video retrieval
- video surveillance
- saliency map
- multimedia
- real time
- key frames
- application layer
- action recognition
- visual features
- input image
- multiresolution
- user interface
- image sequences
- rate adaptation
- computer vision