This paper introduces a novel, to the best of our knowledge, method to estimate and compensate the nonlinear gamma factor introduced by the optical system in fringe projection profilometry. We propose to determine this factor indirectly by adjusting the least-squares plane to the estimated phase coming from the reference plane. We only require a minimal set of three fringe sinusoidal images to estimate the gamma factor. This value can be used to rectify computational legacy data and also to generate and project the new set of fringe patterns for which we perform the inverse gamma compensation. Experimental results demonstrate the feasibility of the proposed method to estimate and correct the gamma distortion.