XVO: Generalized Visual Odometry via Cross-Modal Self-Training.
Lei LaiZhongkai ShangguanJimuyang ZhangEshed Ohn-BarPublished in: ICCV (2023)
Keyphrases
- cross modal
- visual odometry
- multi modal
- autonomous navigation
- long range
- multimedia retrieval
- image retrieval
- kalman filtering
- visual recognition
- depth images
- position information
- ego motion
- visual data
- spatio temporal
- simultaneous localization and mapping
- multimedia
- training set
- dimensionality reduction
- multimedia databases
- pose estimation
- visual similarity
- object recognition
- video sequences
- active learning
- training data