Recently, with the advancement of technologies such as virtual reality and the metaverse, remote 3D reconstruction has gained increasing attention. However, achieving fast and high-quality 3D reconstruction in complex communication environments remains a major challenge. To address this, we propose R-SVRSC, a semantic communication system for single-view 3D reconstruction based on 3D Gaussian Splatting (3D GS). Specifically, the transmitter extracts and transmits semantic information from a single image, instead of transmitting complex 3D data. At the receiver, a Semantic-to-Gaussian Mapper (SGM) directly maps the received deep semantic features into 3D Gaussians, which are then rapidly rendered using a Gaussian rasterizer. Furthermore, to combat the performance degradation caused by multipath fading and channel instability, we introduce a RIS-assisted enhancement to the proposed semantic communication system. A deep learning (DL)-based RIS phase shift prediction module is integrated into the transmitter and jointly trained with the entire system in an end-to-end manner, enabling intelligent reconstruction of the wireless channel. Experimental results show that the proposed system achieves efficient rendering and consistently high-quality 3D reconstruction under low signal-to-noise ratio (SNR) conditions across various real-world channel scenarios.

Reduced-reference (RR) image quality assessment (IQA) metric aims to employ less partial information about the original reference image to achieve higher evaluation accuracy. In this paper, we propose a novel RRIQA metric based on the divisive normalization transform (DNT) in the discrete nonseparable shearlet transform (DNST) domain. In this metric, the coefficients in the DNST domain are normalized employing the Gaussian scale mixture statistical model, and then the marginal distribution of the coefficients changes into approximate Gaussian distribution. A set of statistical features is extracted from DNT-domain representations of the reference and distorted images, respectively. The weighting of these features is performed based on the characteristics of the human visual system. Structural similarity comparison of these features is conducted as an objective quality score of the distorted image. The proposed metric is evaluated on the public LIVE database and demonstrates fairly good performance across a wide range of image distortions.