Purpose: To compare direct clinical and indirect digital photographic assessment of resin composite restorations. Ninety-two posterior resin composite restorations were classified using World Dental Federation (FDI) criteria by two different clinical examiners (C1 and C2). In the same appointment of clinical assessment, intraoral high-quality digital photographs were taken and posteriorly two different digital examiners (D1 and D2) classified the images of each restoration. Restorations of each patient were assessed once by C1 and C2 independently. D1 and D2 assessed the digital images from different locations and in different time. Data were analyzed using the Cohen’s kappa coefficient, Kruskal–Wallis non-parametric test and Dunn's multiple shared test, with 95% confidence. Agreement levels varied from very good (0.81–1.00) to fair (0.21–0.40). Statistically significant differences (p < 0.05) between assessments were found for surface lustre, staining, color match and translucency, esthetic anatomical form, fracture of material and retention and marginal adaptation. The classification of the resin composite restorations varied significantly according to clinical or high-quality digital photographic assessments. Overall, clinical assessment detected more demand for repair or replacement.