The independent study proposed to Jonas Cleveland, ECE Junior, deals with automated 3-D point cloud analysis. Low-level features computation constitutes the first of many processing steps necessary to achieve reliable geometric and semantic scene classification from 3-D laser data.
Jonas's research objectives are to design, implement and test new low level features. The product of this work is to be integrated into an existing software to improve its current performances. During his independent study Jonas will be exposed laser sensing technologies, machine learning algorithms, and geometric methods.
This independent study will be performed within the Vision and Mobile Robotics Laboratory lead by Prof. Martial Hebert. The work will be done in collaboration with Dr. Nicolas Vandapel.