ConvSequential-SLAM: A Sequence-based, Training-less Visual Place Recognition Technique for Changing Environments.
Mihnea-Alexandru TomitaMubariz ZaffarMichael MilfordKlaus D. McDonald-MaierShoaib EhsanPublished in: CoRR (2020)
Keyphrases
- changing environment
- dynamic environments
- autonomous agents
- mobile robot
- visual slam
- visual information
- test set
- visual features
- object and scene recognition
- simultaneous localization and mapping
- training phase
- supervised learning
- training process
- training algorithm
- visual perception
- particle filter
- online learning
- low level
- training set
- e learning
- data sets
- real time