Imaging the near-fault deformation field for large earthquakes has long been a challenging task while it became a critical information for Fault Displacement Hazard Assessment. Classical methods as InSAR and field studies have a hard time grasping the complexity of deformation near the fault zone. Correlation methods using high-resolution optical images appear to be the solution to characterize the displacement field near the fault zone. In this project, we first use very high-resolution (0.5 meters ground resolution) images and a sub-pixel correlation methods (MicMac, IPGP-IGN) to measure the near-fault displacement field associated with large continental earthquakes. We then apply dynamic numerical modeling to our case studies to catch the physics behind surface deformation complexity, like the control on fault geometry, slip distribution and damage pattern. Finally, we will confront all of our results to field observations and have a complete analysis of the different faults systems. First case of studies is the Baluchistan earthquake (2013, Mw 7.7, Pakistan) for which results are very promising and open perspectives on image correlation methods applications.