Traffic Variables Acquisition from Videos

The objective of this project is the development of an automatic tool for the acquisition of vehicular traffic variables, easy to configure. For this 
purpose, different strategies of computer vision (background subtraction, optical flow and edge estimation) are combined for the detection, 
characterization and tracking of the different urban traffic agents.

Scene Flow Estimation from RGBD Images

Scene flow is defined as the motion field in 3D space, and can be computed from a single view when using an RGBD sensor. 
This project addresses the problem of scene flow estimation from RGBD images; by rethinking the way color and depth are exploited.

Local/Global Scene Flow Estimation

Semi-Rigid Scene Flow Estimation