Learning a Depth Covariance Function
Eric Dexheimer, Andrew J. Davison
Abstract
We propose learning a depth covariance function with applications to geometric vision tasks. Given RGB images as input, the covariance function can be flexibly used to define priors over depth functions, predictive distributions given observations, and methods for active point selection. We leverage these techniques for a selection of downstream tasks: depth completion, bundle adjustment, and monocular dense visual odometry.
Topics & Concepts
Artificial intelligenceCovarianceMonocularCovariance functionVisual odometryComputer visionComputer scienceLeverage (statistics)Bundle adjustmentSelection (genetic algorithm)Function (biology)Estimation of covariance matricesMathematicsPattern recognition (psychology)Machine learningImage (mathematics)StatisticsRobotEvolutionary biologyBiologyRobotics and Sensor-Based LocalizationAdvanced Vision and ImagingImage and Object Detection Techniques