Real Time Video Dataset

This research presents a task-oriented and semantics-aware communication framework for augmented reality (TSAR) to enhance communication efficiency and effectiveness in 6G, which includes semantic information extraction, task-oriented semantics-aware wireless communication, avatar pose recovery and rendering. The whole system is based on a targeting machine and a provision machine. The point cloud dataset, produced from FM POINTS, encapsulates the complete Augmented Reality (AR) landscape and written in a PLY file. Originating from a densely packed data set of more than 30,000 individual points, it’s subjected to a process of downsampling, wherein it is reduced to a manageable subset of 2,048 distinct points without significantly compromising the overall visual and data integrity of the AR scenery.