JOKR: Joint Keypoint Representation for Unsupervised Video Retargeting.
Ron MokadyRotem TzabanSagie BenaimAmit H. BermanoDaniel Cohen-OrPublished in: Comput. Graph. Forum (2022)
Keyphrases
- video sequences
- semi supervised
- video streams
- unsupervised manner
- real time
- multimedia
- video representation
- video data
- video content
- keypoints
- unsupervised learning
- digital video
- image representation
- multiscale
- image retargeting
- temporal structure
- content aware
- space time
- video surveillance
- video frames
- image quality
- feature vectors
- image segmentation
- image processing