Abstract: The approach for converting the VIV amplitudes of sectional models to full-scale bridges base on a two-degree-of-freedom empirical model (wake oscillator model) is established in this paper. The wake oscillator model to be used in this paper is firstly introduced. The conversion method by involving fully correlated and non-fully correlated cases are then derived based on this model. A real bridge is utilized as a case study, and the calculated results are compared with those of field measurements. Results indicate that once the span-wise correlation of VIV forces is ignored, the conversion of VIV amplitudes between sectional models and full-scale bridges, generated by wake oscillator model, equals to the result generated by the traditional nonlinear semi-empirical model. However, once the span-wise correlation of VIV forces is considered, the calculated VIV amplitudes of real bridges may be smaller than the results of sectional model tests. The calculated results by ignoring the span-wise correlation of VIV forces agree well with those of field measurements. Thus, the calculation of VIV amplitudes of a full-scale bridge by ignoring the span-wise correlation of VIV forces may be more reliable in the case of lacking an accurate spatial coherence function of VIV forces.