Gwangsoon Lee

3D Gaussian Splatting (3DGS) has recently emerged as an effective representation for immersive 3D scene rendering, providing high visual fidelity and real-time rendering efficiency. To support interoperable compression of trained 3DGS content, the Moving Picture Experts Group (MPEG) is exploring Gaussian Splat Coding (GSC), which mainly targets already trained 3DGS models following the INRIA reference format. The current video-based GSC anchor reorders 3DGS attributes into 2D attribute maps using Parallel Assignment Linear Sorting (PLAS) and compresses the resulting maps using High Efficiency Video Coding (HEVC). However, higher-order spherical harmonic coefficients (SH-AC) often remain irregular and exhibit low local spatial correlation even after PLAS reordering, limiting the coding efficiency of conventional video codecs. This paper proposes a VQ-HEVC hybrid compression framework that is structurally compatible with the video-based GSC anchor framework, in which SH-AC coefficients are represented by vector quantization (VQ) indices, while the remaining attributes are encoded using the same HEVC-based procedure as the GSC anchor. The proposed method adopts a two-stage VQ scheme that combines coarse VQ and product-quantization-based residual quantization, together with zero-masked residual VQ and flexible PQ grouping, to improve index-map coding efficiency across rate points. The generated VQ indices are packed into YUV400 index-map sequences and encoded using HEVC lossless coding, while the corresponding codebooks are transmitted as metadata. Experimental results on the Bartender and Cinema sequences of the MPEG GSC CTC demonstrate consistent rate–distortion improvements over the video-based GSC anchor across multiple objective quality metrics within the evaluated setting. In terms of RGB-PSNR, the proposed method achieves BD-rate reductions of 22.3% and 18.5% for the Bartender and Cinema datasets, respectively. These results suggest that, for the evaluated GSC CTC sequences, VQ-based SH-AC representation can effectively complement PLAS-based video coding while maintaining consistency with the existing GSC coding structure.

In this paper, we introduce the results of providing three-dimensionality and immersion in ordinary UHDTV by realizing the motion parallax from multi-view images in MIV (MPEG Immersive Video) atlas format.

We propose a time-sequential switching light-field (LF) camera for alternative image capture of high-resolution two-dimensional (2D) images and three-dimensional (3D) LF elemental images as additional functionalities. For image data acquisitions of both the 2D and 3D LF imaging of moving objects at a video-frame rate (or even higher frame rate up to approximately 1000 fps), a polarization-dependent-switching micro-lens array (PMLA) is implemented in the LF camera system instead of a conventional passive-type MLA. By controlling the incident polarization conditions using an electrically fast-switching liquid crystal layer, the imaging mode can be time-sequentially switched quite rapidly (switching times of approximately 220 μs and 290 μs for the mode conversions from the 3D LF to the 2D mode, and the reversal mode change, respectively). Using the elemental image sets sampled from the alternating time-sequential imaging results, either directional-view images or depth-refocused images can be reconstructed and provided at a moving picture frame-rate. The depth-refocused images are possible for a wide depth range from 25 cm to 350 cm. Directional views of 22 x 22 and 9 x 9 in portions can be reconstructed for the single shot image capture and the time-sequential video-rate image capture, respectively.

A sub‐pixel multiplexing algorithm for generating multi‐view images on an autostereoscopic three‐dimensional (3D) display at high speed is proposed. Generally, sub‐pixel multiplexing in an autostereoscopic 3DTV system is to assign R, G, B colour sub‐pixels of multiple view images at the proper position on the display such that a proper view image can be seen in accordance with the viewer's position. The upscaled sub‐pixel multiplexing algorithm proposed is composed of bilinear upscaling and a weighted view overlay. Even though upscaling multiple view images requires too much resources and increases system complexity, the proposed algorithm can significantly reduce complexity from an implementation point of view while providing high‐quality 3D views. Specifically, the advantage of the proposed weighted view‐overlay technique is that several overlaid view images decide the colour value of the sub‐pixel, which results in increasing the subjective quality of view images. Through an experiment, it is confirmed that the image quality of generated multi‐views with the proposed algorithms is notably better than the existing simple techniques.