Faculty of Engineering >
Department of Computer Science >
A Novel RRT Extend Function for Efficient and Smooth Mobile Robot Motion Planning
People Detection and Tracking in Sparse 3D Point Clouds
Learning to Detect and Track People in RGB-D Data
Global Localization with FLIRT, Interest Regions for 2D Range Data
With FLIRT features, the problem of global localization can be simply solved using RANSAC. The video shows the global localization result on the Intel data set.
People Tracking in 2D Range Data
AdaBoost In Action
Shows how AdaBoost creates a strong classifier of a toy problem.
Error Ellipses in Action
Shows how the Gaussian error ellipse changes its shape as a function of the covariance matrix, the eigenvalues, and the correlation coefficient.