Urban rooftops, one kind of unused land resource, have become the optimal room for solar photovoltaic deployment. Rooftop solar photovoltaic would be the main source of decarbonizing electricity supply and will play a leading role in the realization of future net-zero carbon cities. An accurate estimation of rooftop solar photovoltaic electrical potential contributes to scientific programming and reasonable arrangement of distributed photovoltaic, land-use efficiency improvement, and ecosystem disturbance reduction. This study aims to comprehensively review rooftop solar photovoltaic potential influencing factors, estimation approaches, and models. The advantages and disadvantages of each approach were discussed, and key directions for future research were summarized. The results show that the rooftop solar photovoltaic electrical potential estimation based on rules of thumb has changed into quantitative and spatial analysis. The adoption of an optimal approach for estimation should trade-off between the assessment scale, accuracy and cost. Among the three current approaches, the sample methodology has lower computational cost and data cost, but uncertainties and less accuracy are the main problems. The complete census methodology has higher evaluation accuracy but is limited by high data acquisition and computing costs. The machine learning method is more advantageous than other approaches for large-scale electrical potential estimation due to its ability in big data mining and algorithm performance improvement. However, there are several problems in the research field, including the gap in accurate large-scale assessment, the uncertainties of outcomes and processes, and the lack of specificity. Three aspects should be highly valued in future studies:1) establish a high-precision simplified model suitable for different regions and complete the technical potential assessment model; 2) figure out the influencing factors of rooftop solar photovoltaic electrical potential, and provide a theoretical basis for the improvement of the representative building classification system and the selection of key feature values; 3) incorporate the impacts of photovoltaic installation scenarios, rural roof quality, urban public building roof ownership and energy demand on building rooftop photovoltaic power potential into the assessment framework.