Predicting Scene Parsing and Motion Dynamics in the Future.
Xiaojie JinHuaxin XiaoXiaohui ShenJimei YangZhe LinYunpeng ChenZequn JieJiashi FengShuicheng YanPublished in: NIPS (2017)
Keyphrases
- image sequences
- image motion
- dynamic scenes
- moving objects
- predicting future
- optical flow fields
- object motion
- consecutive frames
- ground plane
- video sequences
- fixed camera
- space time
- camera motion
- moving camera
- static scenes
- motion model
- visual data
- motion segmentation
- motion estimation
- motion features
- moving observer
- camera images
- three dimensional
- ego motion
- static images
- d scene
- multiple moving objects
- vision sensors
- motion analysis
- motion detection
- motion field
- optical flow
- scene structure
- uncalibrated cameras
- camera movement
- motion parameters
- multiple images
- motion patterns
- affine structure
- point features
- scene understanding
- relative position
- relative depth
- video scene
- physical constraints
- input video
- apparent motion
- flow field
- global motion
- rigid motion
- layered representation
- human motion
- natural language processing
- natural language
- traffic scenes
- multiple objects
- imaging process
- visual motion
- motion capture
- visual scene
- precise localization
- video frames