Abstarct: One of the most important fields in Intelligent Transport Systems is Visual Vehicle Tracking. Main phases are considered tracking and behavior analysis. So, one of them is tracking such as vehicle tracking. Therefore, as tracking is performed, it should be able to obtain accurate tracks from the moment a vehicle enters field of view till it exits. Tracking is challenging when the environment lighting changes, the object changes its size or shape, or occlusion occurs. Tracking at intersection gets more challenging when vehicles change direction at intersections, and more occlusions that happen. That results in losing the tracks. One of the approaches to solve this is using multi-camera systems. As multiple cameras give more information, but also it adds issues such as camera calibration and data association. In this thesis, a new approach for tracking multiple vehicles from a multi-view is proposed to overcome occlusion caused at the intersections with undetermined motion flows. For this purpose, a multilxayer model is presented, which assigns each motion flows to distinct laxyers. Moreover, we introduce different neighborhoods for various laxyers considering the regular motion flows in a laxyer. Hence, vehicles entering from the same side of intersection with the same motion direction are assigned to the same laxyer. Then, a multilxayer graph is presented that assigns motion flows to distinct laxyers with different neighborhoods for each laxyer represented by the graph’s edges. Hence, the vehicle trajectories are distributed among laxyers. Then tracking can be performed in graph of each view separately. After that, all multilxayer graphs of different views are mapped to the graph of the selected view. Then, tracking is performed on the distinct laxyers of the mapped multilxayer graph by computing min-cost flows. In cases such as vehicle crossing, misdetection or occlusion, first the method can predict the vehicle’s tracks by using history, laxyer neighborhoods, and then in graph tracking by using other views’ information. Experimental results show a consistency of the ground truth and the analysis obtained using the proposed approach in tracking vehicles in the inner part of the intersection