Introduction
On this website, we present the UVG-CWI-DQPC dataset, which is a Dual-Quality Point Cloud Dataset for Volumetric Video Applications.
Volumetric video is a key enabler of immersive extended reality (XR) experiences and is often represented using point clouds for their structural simplicity. However, capturing volumetric content through multi-view acquisition and depth sensing poses many challenges, such as occlusions and depth mismatches. To foster research in this field, we introduce a unique dual-quality point cloud dataset, named UVG-CWI-DQPC, which is designed to support the development of point cloud enhancement, compression, and quality assessment. Our dataset includes 12 dynamic sequences captured simultaneously by:
- a high-end capture system producing high-fidelity point clouds with extensive processing; and
- a consumer-grade capture system relying on affordable RGB-D cameras, lightweight processing, and open-source tools.
For each sequence, our dataset provides ground-truth point clouds from the high-end capture system and raw RGB-D footage from the consumer-grade capture system, along with calibration data and tools for point cloud generation. This dual-quality setup enables direct comparison and benchmarking of algorithms for densification, occlusion removal, registration, and quality enhancement. Our dataset is publicly available under a permissive license to support reproducible research and standardization work in Moving Picture Experts Group (MPEG) and 3rd Generation Partnership Project (3GPP).
Contents
The following samples give characteristics of the Dual-Quality Point Cloud sequences in the dataset, with the left image recorded using a high-end and the right image recorded using a consumer grade capturing system:
A person delivers a speech while using hand gestures
Length: 169 frames
Special Features: Moderate movement; Human; Simple textures
A person plays with a volleyball, passing the ball behind the back and between the legs
Length: 171 frames
Special Features: Fast movement; Human; Object; Interactions; Complex texture; High occlusion
A person sits and bounces on a gym ball
Length: 157 frames
Special Features: Global movement; Human; Object; Interactions; Simple textures
A person stretches their back sideways from one side to the other
Length: 201 frames
Special Features: Global movement; Human; Simple textures
A person conducts an eye exam, presenting letters on a chart one at a time
Length: 168 frames
Special Features: Little movement; Human; Objects; Text; Simple textures
A mannequin wearing a head-mounted display (HMD) and a t-shirt with various logos stands still. The t-shirt shifts slightly
Length: 188 frames
Special Features: Static sequence; Objects; Text; Complex textures
A person performs multiple repetitions of the kettlebell swing exercise
Length: 170 frames
Special Features: Global movement; Human; Object; Simple textures
A person adopts various poses facing the camera
Length: 201 frames
Special Features: Little movement; Human; Simple textures
Two persons play Tic Tac Toe using plastic building bricks
Length: 165 frames
Special Features: Fast movement; Humans; Objects; Interactions
A person greets cameras all around with smiles and gestures
Length: 171 frames
Special Features: Fast movement; Human; Simple textures
A person greets the camera with friendly hand gestures and forms a heart shape
Length: 197 frames
Special Features: Moderate movement; Human; Simple textures
A person wearing a head mounted display (HMD) engages in virtual reality gameplay, involving arm movements and body rotations
Length: 196 frames
Special Features: Fast movement; Human; Object; Simple textures
Publications
-
UVG-CWI-DQPC: Dual-Quality Point Cloud Dataset for Volumetric Video Applications}.
In Proceedings of the 33rd ACM International Conference on Multimedia,
2025.
Downloads
-
Dynamic Point Cloud Quality Score & Eye-tracking Dataset (QAVQ-DPC) - UVG-CWI-DQPC: Dual-Quality Point Cloud Dataset for Volumetric Video Applications
LinkDataset
Contact
For questions about this research, please contact Xuemei Zhou (xuemei.zhou [at] cwi.nl)
