"Phase 2: VVC(H.266)-Compressed Dynamic Multiview Volumetric Video Dataset"

"This dataset presents a large-scale collection of VVC (H.266)\u2013compressed multiview dynamic volumetric video content designed for objective and subjective quality assessment in immersive media applications. The dataset is derived from the 8i Voxelized Full Bodies v2 dataset and focuses exclusively on dynamic human performance sequences captured from four viewpoints, enabling realistic evaluation of motion, view dependency, and compression artefacts in volumetric video.The dataset contains 8,448 compressed volumetric video clips, generated from 176 dynamic base sequences, encoded using four quantization parameters (QP = 15, 25, 35, 45) and three compression configurations: geometry-only, texture-only, and combined geometry\u2013texture compression. All content is encoded using the Versatile Video Coding (VVC\/H.266) standard to reflect state-of-the-art compression settings for immersive media delivery.Human subjective quality scores (MOS) collected at QP 25 serve as perceptual ground truth, while the remaining compression levels support predicted MOS (pMOS) generation and cross-QP quality modelling using objective metrics and machine learning approaches. The dataset is organised with a consistent and reproducible folder structure, aligned with the accompanying GitHub repository, to facilitate large-scale benchmarking, reproducible research, and competition-based evaluation.This dataset is intended to support IEEE benchmarking and competition tasks, including volumetric video quality prediction, objective\u2013subjective quality modelling, and compression strategy evaluation for dynamic multiview immersive content."