Exhaust emissions and road runoff pollution pose significant environmental challenges, leading to severe air pollution and irreversible soil damage, respectively. The adoption of self-cleaning road surfaces with photocatalytic properties has emerged as an effective approach to combat both types of pollution. Currently, nano titanium dioxide (TiO2) stands as the most widely used semiconductor for photocatalytic pavements. However, its limited light/radiation utilization remains a concern. To address this, the present study focuses on the use of the green nanomaterial graphene-like phase carbon nitride (g-C3N4), which has a smaller band gap to improve its light utilization. Comprehensive characterizations were performed to elucidate its physical/chemical properties. Subsequently, photocatalytic road coatings with various active substance contents were prepared by incorporating g-C3N4 into epoxy resin, and the functional pavement's performances were evaluated. A laboratory assessing method for the degradation of runoff pollution and NO contaminant was devised, enabling the evaluation of the redox capacity of semiconductive photocatalysts. Experimental results demonstrate that the nano gC3N4 photocatalytic coating exhibits significant enhancement in degradation efficiency for a typical water pollutant, methylene blue (MB), being 2.76 times that of the nano TiO2 photocatalytic coating. Additionally, its ability to degrade NO gaseous contaminant under visible-light irradiation is 7 times more than the control sample of TiO2 coating. Moreover, the photocatalytic coating has good anti-slip property and long-term performance. Thus, nano g-C3N4 proves to be a more suitable semiconductor material for self-cleaning road surfaces, holding tremendous potential to mitigate diverse pollution arising from transportation under natural light conditions.