Fast Monocular Visual-Inertial Initialization Leveraging Learned Single-View Depth.
Nathaniel W. MerrillPatrick GenevaSaimouli KatragaddaChuchu ChenGuoquan HuangPublished in: Robotics: Science and Systems (2023)
Keyphrases
- single view
- multi view
- view invariant
- depth cues
- calibrated cameras
- multiple views
- depth map
- perspective projection
- surface reconstruction
- visual cues
- depth information
- pose estimation
- visual features
- inertial sensors
- polyhedral objects
- ego motion
- co training
- stereo camera
- range images
- viewpoint
- three dimensional
- relative depth
- vanishing points
- d objects
- image sequences