Dual Motion GAN for Future-Flow Embedded Video Prediction.
Xiaodan LiangLisa LeeWei DaiEric P. XingPublished in: CoRR (2017)
Keyphrases
- space time
- object motion
- spatial and temporal
- temporal filtering
- key frames
- input video
- video sequences
- prediction accuracy
- moving camera
- motion features
- predicting future
- optical flow
- temporal continuity
- motion analysis
- image sequences
- image flow
- temporal consistency
- visual data
- video content
- motion prediction
- low frame rate
- motion estimation
- video streams
- motion model
- successive frames
- static images
- video data
- visual cues
- video footage
- shot change detection
- video signals
- flow field
- camera motion
- velocity field
- multimedia
- video scene
- video frames
- human motion
- prediction error
- flow estimation
- dynamic scenes
- temporal coherence
- spatio temporal
- motion trajectories
- layered representation
- moving objects
- dynamic textures
- motion parameters
- camera movement
- motion compensation
- wide area motion imagery
- motion segmentation
- video retrieval
- motion capture data
- single frame
- video analysis
- surveillance videos
- motion capture
- motion patterns
- video summarization
- image motion
- reference frame