In this letter, a combination of characteristic mode (CM) and compressive sensing (CS) is utilized to accelerate the solution of bistatic scattering problems of three-dimensional (3-D) objects. The surface induced currents discretized by Rao–Wilton–Glisson basis functions are successfully sparsely transformed by CM basis functions based on domain decomposition. Different from the dense matrix equation of traditional method of moments (MoM), a low-rank CS model with smaller size is constructed in the proposed method, wherein the induced currents can be reconstructed efficiently. Due to partially filling of the impedance matrix and efficient recovery algorithm, the time cost in the proposed method is reduced dramatically. Last, the results from several numerical simulations validate the efficiency and accuracy of the proposed method.