Proper building maintenance is critical to ensuring the well-being of citizens and sustainable development of the urban environment. For high-rise high-density cities, one challenge to building maintenance concerns the inspection of building exteriors, which has been mostly done manually but is of highly risky, labor-intensive, and error-prone. Unmanned aerial vehicles (UAVs) have the potential to support building exterior inspection in a safe and efficient way. However, the effective application of UAVs for exterior inspection is hindered by path planning problems in high-rise high-density urban areas. Previously developed path planning algorithms are constrained by algorithmic complexity and cannot be implemented in a large three-dimensional space. Besides, path planning assisting tools available in the market that supported by Google Earth Pro can hardly assure high accurate results. In this paper, a new virtual prototyping-based path planning system is developed for UAVs for building exterior inspection in high-rise high-density urban areas. Unreal Engine as a widely-used game engine was applied. The system provides a realistic game world to simulate real-world activities, which enables operators to design UAV paths like playing games. The flight paths and viewpoints can be repeatedly tested in a gaming context until requirements are satisfied. The system integrates expert knowledge through "human in the loop" and realistic information on the virtual environment featured with repeatability. The system thereby greatly alleviates the risks that are induced by physical constraints in the real world when adopting UAV, and thus should significantly improve the effectiveness and efficiency of building exterior inspection.