This paper presents a method for multi-camera people tracking using a probabilistic occupancy map and dynamic programming. The authors demonstrate that their approach can effectively handle occlusions and lighting changes, accurately tracking up to six individuals across thousands of frames. The method combines a generative model that estimates the probabilities of occupancy of the ground plane with a color and motion model to track individuals over time. The generative model uses simple rectangles to create synthetic images, minimizing the Kullback-Leibler divergence from the true posterior distribution. This allows for the computation of occupancy probabilities as the fixed point of a system of equations. The paper also introduces a greedy algorithm to process individual trajectories separately, ensuring robustness in challenging situations. Experimental results show that the method performs well, with accurate location estimates and high robustness to image losses.This paper presents a method for multi-camera people tracking using a probabilistic occupancy map and dynamic programming. The authors demonstrate that their approach can effectively handle occlusions and lighting changes, accurately tracking up to six individuals across thousands of frames. The method combines a generative model that estimates the probabilities of occupancy of the ground plane with a color and motion model to track individuals over time. The generative model uses simple rectangles to create synthetic images, minimizing the Kullback-Leibler divergence from the true posterior distribution. This allows for the computation of occupancy probabilities as the fixed point of a system of equations. The paper also introduces a greedy algorithm to process individual trajectories separately, ensuring robustness in challenging situations. Experimental results show that the method performs well, with accurate location estimates and high robustness to image losses.