live ut

Laboratory for Image & Video Engineering

Welcome to the CHUG HDR UGC Video Quality Dataset

CHUG: Crowdsourced User‑Generated HDR Video Quality Dataset

Download / Access Description Examples Citation

Introduction

High Dynamic Range (HDR) videos offer superior brightness, contrast, and color volume. With the surge of user‑generated content (UGC) on modern platforms, assessing HDR video quality in the wild has become increasingly important. We present CHUG, a large‑scale HDR-UGC video quality dataset created via a controlled crowdsourcing study. The dataset contains 856 HDR-UGC source videos, transcoded across multiple resolutions and bitrates to simulate realistic delivery conditions, resulting in a total of 5,992 videos. A large‑scale subjective study collected 211,848 quality ratings. We analyze reliability and behavior of human responses and benchmark existing VQA models, highlighting significant room for improvement on HDR-UGC content.

Exemplar frames from CHUG
Sample frames from the CHUG dataset, with diverse lighting, motion, orientation, and distortions. Best viewed zoomed in.
Download

We are making the CHUG HDR-UGC Video Quality Dataset available to the research community. If you use this dataset in your research, we kindly ask that you cite our paper and this website as listed below.

  • S. Saini, A. C. Bovik, N. Birkbeck, Y. Wang, B. Adsumilli, "CHUG: Crowdsourced User‑Generated HDR Video Quality Dataset," 2025 IEEE International Conference on Image Processing (ICIP), Anchorage, AK, USA, 2025, pp. 2504‑2509, doi: 10.1109/ICIP55913.2025.11084488.
  • S. Saini, A. C. Bovik, N. Birkbeck, Y. Wang and B. Adsumilli, "CHUG: Crowdsourced User-Generated HDR Video Quality Dataset", Online: https://live.ece.utexas.edu/research/chug/index.html, 2025.

Access Instructions

Metadata and step‑by‑step instructions are available in the GitHub repository. Example commands to fetch videos using AWS CLI are shown below.

chug.csv (IDs) 

# Download a single video by ID
aws s3 cp s3://ugchdrmturk/videos/VIDEO_ID.mp4 ./CHUG_Videos/

# Download all listed video IDs from chug.csv or chug-video.txt
# (ensure one ID per line)
cat chug-video.txt | while read video; do
  aws s3 cp s3://ugchdrmturk/videos/${video}.mp4 ./CHUG_Videos/
done

You can also play a video directly by replacing VIDEO_ID in this URL:

https://ugchdrmturk.s3.us-east-2.amazonaws.com/videos/VIDEO_ID.mp4

Example: 9ae245a27cc5ea9d2f3fae9692250281.mp4

Examples

A few example clips. Videos are HDR; for best results, view on an HDR-capable display.

Sample Videos (Portraits)
Indoor Scene
Museum
Costumes
Sample Videos (Landscapes)
Carousel
Nature
Sunset
Screen
Light Show
Birds
Database Description

The CHUG Database contains 5,992 video sequences generated from 856 real‑world UGC‑HDR source videos. Sources were transcoded across multiple resolution–bitrate ladders (e.g., 144p–1080p; a range of practical bitrates) to create realistic delivery conditions. All videos were evaluated via a controlled large‑scale crowdsourcing study, yielding 211,848 quality ratings. We benchmark a variety of existing VQA models on their ability to predict UGC‑HDR quality and discuss headroom for progress on this modality.

Investigators

Copyright Notice & Citation

-----------COPYRIGHT NOTICE STARTS WITH THIS LINE------------
Copyright (c) 2025 The University of Texas at Austin
All rights reserved.

Permission is hereby granted, without written agreement and without license or royalty fees, to use, copy, modify, and distribute this database (the videos, the results and the source files) and its documentation for any purpose, provided that the copyright notice in its entirety appear in all copies of this database, and the original source of this database, Laboratory for Image and Video Engineering (LIVE, http://live.ece.utexas.edu/) at the University of Texas at Austin (UT Austin, http://www.utexas.edu/), is acknowledged in any publication that reports research using this database.

The following paper/website are to be cited in the bibliography whenever the database is used as:

  • S. Saini, A. C. Bovik, N. Birkbeck, Y. Wang and B. Adsumilli, "CHUG: Crowdsourced User-Generated HDR Video Quality Dataset," 2025 IEEE International Conference on Image Processing (ICIP), Anchorage, AK, USA, 2025, pp. 2504-2509, doi: 10.1109/ICIP55913.2025.11084488.
  • S. Saini, A. C. Bovik, N. Birkbeck, Y. Wang and B. Adsumilli, "CHUG: Crowdsourced User-Generated HDR Video Quality Dataset", Online: https://live.ece.utexas.edu/research/chug/index.html, 2025.

IN NO EVENT SHALL THE UNIVERSITY OF TEXAS AT AUSTIN BE LIABLE TO ANY PARTY FOR DIRECT, INDIRECT, SPECIAL, INCIDENTAL, OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE USE OF THIS DATABASE AND ITS DOCUMENTATION, EVEN IF THE UNIVERSITY OF TEXAS AT AUSTIN HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.

THE UNIVERSITY OF TEXAS AT AUSTIN SPECIFICALLY DISCLAIMS ANY WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE. THE DATABASE PROVIDED HEREUNDER IS ON AN "AS IS" BASIS, AND THE UNIVERSITY OF TEXAS AT AUSTIN HAS NO OBLIGATION TO PROVIDE MAINTENANCE, SUPPORT, UPDATES, ENHANCEMENTS, OR MODIFICATIONS.

-----------COPYRIGHT NOTICE ENDS WITH THIS LINE------------

Back to Quality Assessment Research page