Unsupervised Selection of Optimal Operating Parameters for Visual Place Recognition Algorithms Using Gaussian Mixture Models.
James MountMing XuLes DawesMichael MilfordPublished in: IEEE Robotics Autom. Lett. (2021)
Keyphrases
- significant improvement
- parameter tuning
- least squares
- worst case
- neural network
- computationally efficient
- data mining
- optimal parameters
- dynamic programming
- optimization problems
- vantage points
- approximately optimal
- fine tuning
- visual perception
- image classification
- supervised learning
- computational cost
- data structure
- face recognition
- high level
- decision trees
- machine learning