Population size is a highly important parameter for wildlife management and conservation. However, its estimation can be challenging when a portion of the population is undetectable. Spatially explicit capture-recapture models are a precise approach to estimate wildlife population density while accounting for imperfect detection, but all animals must be individually identifiable. Spatial mark-resight models (SMR) allow the estimation of population sizes when only some individuals can be identified. This is the case in feral pigs (Sus scrofa), where some individuals are recognizable by natural marks. We compared two SMR approaches to estimate feral pig population density: SMR for an unknown number of marked individuals (SMR-UM) and SMR for a known number of marked individuals (SMR-KM). Both approaches are applicable in species with some individuals recognisable by natural marks, such as feral pigs. The SMR-KM is similar to a process of capture-mark-recapture of fewer individuals, which can be used in species with non-recognisable individuals (e.g., wild boar, S. scrofa). First, we fitted a SMR-UM using the complete capture history (individuals/traps/days) for all recognisable individuals (n=33) and the latent capture history (traps-days) for unmarked individuals throughout the entire sampling occasion (66 days). Secondly, we fitted SMR-KM dividing the sampling occasions into two periods: the sighting period (25 days) to identify individuals (n=13), and the resighting period (41 days) in which we used the complete capture and latent capture histories of the marked and unmarked individuals, respectively. We estimated very similar densities with the two approaches for feral pigs in our study area: 13.27 (SD=3.07) (8.12-20.02 95\% BCI) and 12.87 (SD=2.21) (8.96-17.59 95\% BCI) pigs/km², for SMR-UM and SMR-KM, respectively. Our results indicate that SMR models are an effective tool to monitor feral pig populations, as well as similar non-individually identifiable species, by tagging a small sample of the population.