MVFuseNet: Improving End-to-End Object Detection and Motion Forecasting through Multi-View Fusion of LiDAR Data.
Ankit LaddhaShivam GautamStefan PalomboShreyash PandeyCarlos Vallespi-GonzalezPublished in: CoRR (2021)
Keyphrases
- multi view
- end to end
- lidar data
- object detection
- dynamic scenes
- multiple views
- depth map
- three dimensional
- image sequences
- urban environments
- semi supervised
- d objects
- high resolution
- optical flow
- camera motion
- motion estimation
- face detection
- point cloud
- urban areas
- range images
- multi sensor
- moving objects
- feature points
- motion vectors
- background subtraction
- high quality
- video surveillance
- image quality
- computer vision