Land surface albedo is an important variable for Earth’s radiation and energy budget. Over the past decades, many surface albedo products have been derived from a variety of remote sensing data. However, the estimation accuracy, temporal resolution, and temporal continuity of these datasets still need to be improved. We developed a multi-sensor strategy (MSS) based on the direct-estimation algorithm (DEA) and Statistical-Based Temporal Filter (STF) to improve the quality of land surface albedo datasets. The moderate-resolution imaging spectroradiometer (MODIS) data onboard Terra and Aqua and the Visible Infrared Imaging Radiometer Suite (VIIRS) onboard the Suomi-National Polar-orbiting Partnership (NPP) were used as multi-sensor data. The MCD43A3 product and in situ measurements from the Surface Radiation Budget Network (SURFRAD) and FLUXNET sites were employed for validation and comparison. The results showed that the proposed MSS method significantly improved the temporal continuity and estimation accuracy during the snow-covered period, which was more consistent with the measurements of SURFRAD (R = 0.9498, root mean square error (RMSE) = 0.0387, and bias = −0.0017) and FLUXNET (R = 0.9421, RMSE = 0.0330, and bias = 0.0002) sites. Moreover, this is a promising method to generate long-term, spatiotemporal continuous land surface albedo datasets with high temporal resolution.