Optical flow-based segmentation of moving objects for mobile robot navigation using pre-trained deep learning models.
Artúr István KárolyRenáta Nagyné ElekTamás HaideggerKároly SzéllPéter GalambosPublished in: SMC (2019)
Keyphrases
- learning models
- mobile robot navigation
- pre trained
- optical flow
- mobile robot
- training data
- learning algorithm
- training examples
- loss function
- unknown environments
- learning tasks
- image sequences
- machine learning
- semi supervised learning
- machine learning algorithms
- classification models
- motion parameters
- motion estimation
- conditional random fields
- flow field
- control signals
- topological map
- motion model
- d scene
- learning problems
- decision trees
- reinforcement learning
- camera motion
- neural network
- active learning
- higher order
- dynamic environments