Soilwater erosion is a serious problem, especially in agricultural lands. Among these, vineyards deserve attention, because they constitute for the Mediterranean areas a type of land use affected by high soil losses. A significant problem related to the study of soil water erosion in these areas consists in the lack of a standardized procedure of collecting data and reporting results, mainly due to a variability among the measurement methods applied. Given this issue and the seriousness of soilwater erosion inMediterranean vineyards, this works aims to quantify the soil losses caused by simulated rainstorms, and compare them with each other depending on two different methodologies: (i) rainfall simulation and (ii) surface elevation change-based, relying on high-resolution Digital Elevation Models (DEMs) derived from a photogrammetric technique (Structure-from-Motion or SfM). The experiments were carried out in a typical Mediterranean vineyard, located in eastern Spain, at very fine scales. SfMdatawere obtained fromone reflex camera and a smartphone built-in camera. An index of sediment connectivity was also applied to evaluate the potential effect of connectivity within the plots. DEMs derived from the smartphone and the reflex camera were comparable with each other in terms of accuracy and capability of estimating soil loss. Furthermore, soil loss estimatedwith the surface elevation change-basedmethod resulted to be of the same order ofmagnitude of that one obtained with rainfall simulation, as long as the sediment connectivity within the plotwas considered.High-resolution topography derived fromSfMrevealed to be essential in the sediment connectivity analysis and, therefore, in the estimation of eroded materials, when comparing themto those derived from the rainfall simulation methodology. The fact that smartphones built-in cameras could produce as much satisfying results as those derived from reflex cameras is a high value added for using SfM.