Oral title: First statistical constraints from the SHINE direct imaging survey Oral abstract: The SHINE survey with SPHERE aims at providing the best statistical constraints to date on the population of giant exoplanets by observing a broad sample of 400 to 600 young, nearby stars. The high-contrast and high-angular resolution capabilities offered by SPHERE will provide for the first time a reliable estimation of the giant planets frequency in the 5-100 AU range and down to a few Jupiter masses. After briefly presenting the status of the observations and the classification of the 2500+ candidates, I will present the first statistical constraints derived from the observations of more than 160 stars. These first results are based on two complementary approaches. The first phenomenological approach parameterizes the observed planet mass function (as a constant power law or as depending on host star mass) and orbital surface density distribution (with radial power-law and log-normal) to constrain model parameters compared to data. The second one relies on population synthesis models based on both gravitational instability and core accretion formation scenarios. For the first time direct imaging observations start to provide meaningful constraints on giant planet formation models. Finally, I will conclude by presenting the prospects for the statistical analysis of the complete SHINE sample.