Indoor navigation is a challenging problem that requires intelligent systems to guide users through complex and dynamic environments. The conventional Global Positioning System (GPS) is not suitable for indoor navigation, as it relies on satellite signals that are often blocked by architectural structures. Therefore, there is a growing demand for effective indoor navigation solutions that can cope with the intricacies of modern urban landscapes, where buildings are designed with multiple floors, rooms, and corridors. This paper introduces a state-of-the-art augmented reality indoor navigation application, which seamlessly integrates computer vision technology, device sensors, and an optimized A-star algorithm. The main advantage of this system is that it uses computer vision to comprehensively analyze and interpret the indoor environment, capturing spatial features and dynamically updating a real-time 3D map. This holistic understanding enables precise user localization and augments the overall navigation experience with visual cues and annotations. The integration of device sensors, such as accelerometer, gyroscope, and magnetometer, ensures accurate orientation-based directions as users traverse indoor spaces, enhancing the utility of the application. The enhanced A-star algorithm not only optimizes route planning for efficiency, but also incorporates approximation methods to strategically manage device resource consumption during route estimation. This amalgamation of cutting-edge technologies not only addresses the challenges posed by intricate interior designs, but also establishes a versatile and resource-efficient means of guiding users through complex indoor environments. The vision-based intelligent system presented herein stands as a beacon in revolutionizing indoor navigation, particularly in the context of modern urban landscapes.