MGTANet: Encoding Sequential LiDAR Points Using Long Short-Term Motion-Guided Temporal Attention for 3D Object Detection.
Junho KohJunhyung LeeYoungwoo LeeJaekyum KimJun Won ChoiPublished in: CoRR (2022)
Keyphrases
- short term
- d objects
- long term
- motion prediction
- lidar data
- space time
- geometric invariants
- viewing direction
- physically plausible
- pose parameters
- multi view
- viewpoint
- short term and long term
- object recognition
- surface points
- feature points
- pose estimation
- long term memory
- range data
- three dimensional
- cad models
- d mesh
- image sequences
- multiple views
- motion estimation
- forecasting model
- load forecasting
- three dimensional objects
- object contours
- medium term
- machine learning
- shape descriptors
- point cloud
- pose normalization
- point distribution
- object detection
- object classes
- optical flow
- projective geometry
- surface properties
- short and long term
- multi views
- focus of attention
- urban environments
- point sets
- integral imaging
- arima model
- minimum bounding
- visual attention
- video coding
- video sequences