Shear wave motion detection is a critical part of ultrasound shear wave elastography (SWE). Shear wave signal-to-noise-ratio (SNR) is strongly related to the ultrasound RF (radiofrequency) signal SNR. Recently, we demonstrated substantial improvement of shear wave motion detection using pulse-inversion harmonic imaging (PIHI). In this study we propose to use filter-based harmonic imaging (FHI) to realize dual-frequency shear wave motion detection. With a filter-based approach, one can calculate shear wave motion signal from both the fundamental component and the harmonic component of the RF signal that is acquired from a single pulse-echo cycle. Two types of common shear wave imaging methods were investigated using dual-frequency detection: shear wave imaging with acoustic radiation force (ARF) as the shear wave source, and shear wave imaging with external mechanical vibration as the shear wave source. Phantom studies showed that for ARF based SWE, the push beam attenuates faster than the harmonic signal, and therefore, one can use harmonic signal for shear wave detection throughout the range of depth; for external mechanical vibration-based SWE, one can use harmonic signal to calculate shear wave motion for shallower region of the tissue for better SNR, and fundamental signal to calculate shear wave motion for deeper region of the tissue for better penetration, so that robust shear wave motion detection can be achieved throughout a large range-of-depth.