This research looks into whether it is possible to use SNNs along with deep learning methods to find occlusions in virtual reality and augmented reality images. The next step goes into more detail about the basic ideas behind SNNs and the benefits they offer, such as event-driven processing and low power usage, which are both very important for real-time augmented and virtual reality systems. After that, we’ll talk about our one-of-a-kind occlusion recognition system, which uses both deep learning and SNNs. Utilizing both virtual and real-world AR and VR datasets, we conduct experiments to test how well our method works. These results show a big improvement in the accuracy of occlusion recognition compared to previous methods. We also find out how well our system works with computers and how many resources it needs. This shows that it can be used on AR and VR devices that don’t have a lot of resources. In the end of this research, it is shown that spiking neural networks and deep learning methods can make it easier to find occlusions in AR/VR pictures. Our method improves augmented and virtual reality experiences by getting rid of this major problem. This opens up new possibilities in many areas, such as education, training, simulations, and gaming.