Unified Multi-Modal Landmark Tracking for Tightly Coupled Lidar-Visual-Inertial Odometry.
David WisthMarco CamurriSandipan DasMaurice F. FallonPublished in: IEEE Robotics Autom. Lett. (2021)
Keyphrases
- multi modal
- tightly coupled
- inertial sensors
- sensor fusion
- cross modal
- fine grained
- position and orientation
- motion tracking
- loosely coupled
- landmark recognition
- general purpose
- video search
- multi modality
- audio visual
- kalman filter
- visual information
- position estimation
- single modality
- particle filter
- high dimensional
- point cloud
- low level
- multiple modalities
- auto annotation
- fusing multiple
- high level
- semantic concepts
- mean shift
- image registration
- high resolution
- image sequences
- humanoid robot
- image annotation
- distributed systems