Abstract: Light-field particle image velocimetry (LF-PIV) was recently introduced to measure three-dimensional, three-component velocity field with just a single light-field camera. One of the major challenges lies in the small viewing aperture affecting the depth resolution of such a single-camera based LF-PIV approach. In the present study, we show that this limitation may be mitigated by a dual light-field camera framework, one which includes a novel volumetric calibration model derived from Gaussian optics, a particle intensity reconstruction algorithm based on the multiplicative algebraic reconstruction technique and a post-processing technique for the reconstructed particle intensity field. The proposed approach was firstly validated with synthetic light-field particle images as well as experimental light-field images of five tiny glass beads imitating tracer particles. Secondly, parametric studies were conducted to analyze the influence of the viewing angle and seeding particle density on the reconstruction quality and spatial resolution. In particular, synthetic light-field particle images of a direct numerical simulation jet data set were utilized to compare the performance of single- and dual-camera LF-PIV techniques. Finally, experimental volumetric flow field results of a circular vortex-ring were also measured by single- and dual-camera LF-PIV techniques and compared. It is determined here that an additional light-field camera can mitigate the elongation effects of reconstructed particles and improve the measurement resolution in the depth direction. Graphic abstract: [Figure not available: see fulltext.].