The idea of this project is to create a base-level 3D model from satellite imagery by means of photogrammetry and machine learning, then improve its resolution by superimposing data from other sources (such as aerial footage, drone footage, street view footage, and Lidar data). The process applies advanced algorithms like S2P, which is an open-source software for structure from motion that is specifically designed for satellite imagery data. By using satellite imagery data, more accurate 3D models can be obtained and refined. The process also involves taking multiple images of the same area from different angles and positions, with the help of a satellite. The proposed research is looking into a hybrid model to increase 3d model resolution and building textures.