Medical Point Clouds Enhancement at the Network Edge

The paper proposes a method for the enhancement of medical point clouds suitable for implementation at the edge of a next generation network. This method exploits the 2D point clouds projection employed in compression algorithms and enhances the point cloud by applying a diffusion sampling model in the flattened domain. The proposed approach, to which we refer to as Projection Sampling based Point Cloud Enhancement, perfectly fits the e-health service architecture in next generation network since it can be implemented at the network edge or within the network, at an intermediate transcoding stage. The experimental findings with medical point clouds demonstrate the method’s efficacy in mitigating noise and preserving texture information, making it a valuable tool for incorporating point cloud enhancement into an Extended Reality transmission system.