We propose a new general SLAM system that uses the semantic segmentation of objects and structures in the scene. Semantic information is relevant as it contains high level information which may make SLAM more accurate and robust. Our contribution is threefold: i) A new SLAM system based on ORB-SLAM2 that creates a semantic map made of clusters of points corresponding to objects instances and structures in the scene. ii) A modification of the classical Bundle Adjustment formulation to constrain each cluster using geometrical priors, which improves both camera localization and reconstruction and enables a better understanding of the scene. iii) A new Bundle Adjustment formulation at the level of clusters to improve the convergence of classical Bundle Adjustment. We evaluate our approach on several sequences from a public dataset and show that, with respect to ORB-SLAM2 it improves camera pose estimation.