Resumo. Due to the increasing environmental degradation, the understanding of how reforestation and other restoration techniques function became fundamental to base the reestablishment of degraded ecosystems. The aim of this study was to perform a vegetation survey in a restoration site through three different techniques. The study site has 7.2 ha in a subtropical forest in Dois Vizinhos, Paraná state, Brazil and it is approximately 20 m from a secondary forest fragment. The experiment consisted of 12 plots with three treatments (T1-3) randomly distributed in four blocks. T1: natural regeneration through passive restoration. T2: set of nucleation techniques. T3: seedlings planted through filling and diversity lines (3x2 m). For sampling purposes, each plot was divided into 24 subplots. In order to evaluate plant presence, two years and eight months after the treatments implementation, a survey was conducted in three random subplots for each plot, where all plants taller than 50cm were identified and quantified. Native species richness was higher than 87% in all treatments, whereas the presence of herbaceous species was higher at passive restoration and woody species at plantation. The most common dispersal syndromes were anemochory and zoochory. By comparing planted and regenerated individuals for the three treatments, the plantation technique had the highest Shannon diversity index (3.017), with 148 individuals and 53 species sampled. However, there was no statistically significant difference. This higher diversity is related to the larger number of species planted, but it is still not facilitating regeneration species in its understory at this age. Considering only regenerated individuals, nucleation showed the highest Shannon diversity index (2.078) with a total of 908 individuals of 40 species, however, it wasn’t statistically significant different of passive restoration (Shannon = 1.965) which had 2,018 individuals of 40 species. The most floristic similar treatments were passive restoration and nucleation. Nucleation facilitated species natural regeneration (17 planted and 40 regenerated species) while passive restoration showed high resilience, therefore the use of passive restoration can be considered an effective alternative since its results did not differ statistically from nucleation, and the main advantage of this technique is the reduced cost.