Filters








116 Hits in 4.8 sec

YouTube UGC Dataset for Video Compression Research [article]

Yilin Wang, Sasi Inguva, Balu Adsumilli
2019 arXiv   pre-print
Understanding difficulties for compression and quality assessment in the scenario of UGC is important, but there are few public UGC datasets available for research.  ...  This paper introduces a large scale UGC dataset (1500 20 sec video clips) sampled from millions of YouTube videos.  ...  CONCLUSION This paper introduced a large scale dataset for UGC videos, which highly represents the videos uploaded to YouTube.  ... 
arXiv:1904.06457v2 fatcat:suek2m5x3bdm5bj66ipkupcb3y

Subjective Quality Assessment for YouTube UGC Dataset [article]

Joong Gon Yim, Yilin Wang, Neil Birkbeck, Balu Adsumilli
2020 arXiv   pre-print
To facilitate compression-related research on UGC, YouTube has released a large-scale dataset. The initial dataset only provided videos, limiting its use in quality assessment.  ...  We used a crowd-sourcing platform to collect subjective quality scores for this dataset.  ...  To enable novel research on compression and quality assessment on UGC, in this work we have collected and publicly released the corresponding subjective quality scores for the YouTube UGC dataset 1 .  ... 
arXiv:2002.12275v1 fatcat:pyia6mi5rfhbjn6oeeuxcmupoq

UGC-VQA: Benchmarking Blind Video Quality Assessment for User Generated Content [article]

Zhengzhong Tu, Yilin Wang, Neil Birkbeck, Balu Adsumilli, Alan C. Bovik
2021 arXiv   pre-print
Accordingly, there is a great need for accurate video quality assessment (VQA) models for UGC/consumer videos to monitor, control, and optimize this vast content.  ...  Our study protocol also defines a reliable benchmark for the UGC-VQA problem, which we believe will facilitate further research on deep learning-based VQA modeling, as well as perceptually-optimized efficient  ...  The most recently published UGC-VQA database is the YouTube-UGC Dataset [11] comprising 1,380 20-second video clips sampled from millions of YouTube videos, which were rated by more than 8,000 human  ... 
arXiv:2005.14354v2 fatcat:bjixvu7ryza4vglncr6v23eal4

Compression of user generated content using denoised references [article]

Eduardo Pavez, Enrique Perez, Xin Xiong, Antonio Ortega, Balu Adsumilli
2022 arXiv   pre-print
We demonstrate the effectiveness of the proposed strategy for JPEG compression of UGC images and videos.  ...  Video shared over the internet is commonly referred to as user generated content (UGC). UGC video may have low quality due to various factors including previous compression.  ...  Experiments with YouTube UGC dataset YouTube UGC is a large scale dataset sampled from Youtube videos.  ... 
arXiv:2203.03553v2 fatcat:ronfee7ggjfo3mf7bsrqpldvki

A strong baseline for image and video quality assessment [article]

Shaoguo Wen, Junle Wang
2021 arXiv   pre-print
Based on the architecture proposed, we release the models well trained for three common real-world scenarios: UGC videos in the wild, PGC videos with compression, Game videos with compression.  ...  In this work, we present a simple yet effective unified model for perceptual quality assessment of image and video.  ...  ., LIVE-VQC [10] , KoNViD-1K [11] and YouTube-UGC [12] , and three private datasets in different commercial scenarios, i.e., UGC videos in the wild, PGC videos with compression and Game videos with  ... 
arXiv:2111.07104v1 fatcat:fdfgal5j4rb2vkpxcuwz3teql4

RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content

Zhengzhong Tu, Xiangxu Yu, Yilin Wang, Neil Birkbeck, Balu Adsumilli, Alan C. Bovik
2021 IEEE Open Journal of Signal Processing  
Our experimental results on recent large-scale UGC video quality databases show that RAPIQUE delivers top performances on all the datasets at a considerably lower computational expense.  ...  Accurate and efficient video quality predictors suitable for this content are thus in great demand to achieve more intelligent analysis and processing of UGC videos.  ...  After graduation, he joined the Media Algorithm team in Youtube/Google. His research fields include video processing infrastructure, video quality assessment, and video compression.  ... 
doi:10.1109/ojsp.2021.3090333 fatcat:7yn553wfzzfk5j2jsdgwwvsr3m

Perceptual Quality Assessment of UGC Gaming Videos [article]

Xiangxu Yu, Zhengzhong Tu, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
2022 arXiv   pre-print
specifically designed for gaming videos.  ...  However, relatively little research has been done on the automatic quality prediction of gaming videos, especially on those that fall in the category of "User-Generated-Content" (UGC).  ...  ACKNOWLEDGMENT This research was supported by a gift from YouTube, and by grant number 2019844 from the National Science Foundation AI Institute for Foundations of Machine Learning (IFML).  ... 
arXiv:2204.00128v2 fatcat:bv2zwkuu7bau5dlyb75bhgybjq

RAPIQUE: Rapid and Accurate Video Quality Prediction of User Generated Content [article]

Zhengzhong Tu, Xiangxu Yu, Yilin Wang, Neil Birkbeck, Balu Adsumilli, Alan C. Bovik
2021 arXiv   pre-print
Our experimental results on recent large-scale UGC video quality databases show that RAPIQUE delivers top performances on all the datasets at a considerably lower computational expense.  ...  Accurate and efficient video quality predictors suitable for this content are thus in great demand to achieve more intelligent analysis and processing of UGC videos.  ...  -1k and YouTube-UGC.  ... 
arXiv:2101.10955v1 fatcat:sg4xf6rg3rhy3pnc3yx5az2joy

CONVIQT: Contrastive Video Quality Estimator [article]

Pavan C. Madhusudana and Neil Birkbeck and Yilin Wang and Balu Adsumilli and Alan C. Bovik
2022 arXiv   pre-print
Perceptual video quality assessment (VQA) is an integral component of many streaming and video sharing platforms.  ...  Here we consider the problem of learning perceptually relevant video quality representations in a self-supervised manner.  ...  The authors would also like to acknowledge the Texas Advanced Computing Center (TACC) for providing computational resources that contributed to this research, and YouTube for supporting this work.  ... 
arXiv:2206.14713v1 fatcat:boy77ymb5rbbxadjmcygfoyon4

A Deep Learning based No-reference Quality Assessment Model for UGC Videos [article]

Wei Sun, Xiongkuo Min, Wei Lu, Guangtao Zhai
2022 arXiv   pre-print
Quality assessment for User Generated Content (UGC) videos plays an important role in ensuring the viewing experience of end-users.  ...  Previous UGC video quality assessment (VQA) studies either use the image recognition model or the image quality assessment (IQA) models to extract frame-level features of UGC videos for quality regression  ...  For KoNViD-1k, YouTube-UGC, LBVD, and LIVE-YT-Gaming databases, we randomly split these databases into the training set with 80% videos and the test set with 20% videos for 10 times, and report the median  ... 
arXiv:2204.14047v1 fatcat:2r6j3sndorblbbmex2mawch3bq

VMAF And Variants: Towards A Unified VQA [article]

Pankaj Topiwala, Wei Dai, Jiangfeng Pian, Katalina Biondi, Arvind Krovvidi
2021 arXiv   pre-print
When fully trained, FR algorithms such as VMAF perform very well on test datasets, reaching 90%+ match in PCC and SRCC; but for predicting performance in the wild, we train/test from scratch for each database  ...  We investigate variants of the popular VMAF video quality assessment algorithm for the FR case, using both support vector regression and feedforward neural networks.  ...  Fig. 4 gives example images from FR datasets BVIHD [14] , NFLX-2 [4] , and the NR dataset YouTube UGC [15] . Note that the BVIHD dataset is challenging for VMAF; see Fig. 5 .  ... 
arXiv:2103.07770v7 fatcat:4zeezqncxvcnbkv3o3a5lk36hq

Per-clip and per-bitrate adaptation of the Lagrangian multiplier in video coding [article]

Daniel J. Ringis, François Pitié, Anil Kokaram
2022 arXiv   pre-print
In the experiments presented we employ direct optimization techniques to estimate this Lagrangian parameter path approximately 2,000 video clips. The clips are primarily from the YouTube-UGC dataset.  ...  In our recent work we have presented per-clip optimization for the Lagrangian multiplier in Rate controlled compression, which yielded BD-Rate improvements of approximately 2\% across a corpus of videos  ...  ACKNOWLEDGMENTS This work was supported in part by YouTube, Google and the Ussher Research Studentship from Trinity College.  ... 
arXiv:2204.09055v1 fatcat:efusggfjljbgnmuumkpv664d5m

Multiview Contrastive Learning for Completely Blind Video Quality Assessment of User Generated Content [article]

Shankhanil Mitra, Rajiv Soundararajan
2022 arXiv   pre-print
The design of this class of methods is particularly important since it can allow for superior generalization in performance across various datasets.  ...  We consider the design of completely blind VQA for user generated content.  ...  Though 𝑉𝐶𝑂𝑅𝑁 𝐼𝐴 * (1fr/sec), and pretrained 𝑅𝑒𝑠𝑁 𝑒𝑡50 * (1 fr/sec) performs well on the Youtube-UGC [40] dataset, its performance on other datasets is poor.  ... 
arXiv:2207.06148v1 fatcat:zjnzcm657ngejpriauwtz2t5wu

Enhancing VVC with Deep Learning based Multi-Frame Post-Processing [article]

Duolikun Danier, Chen Feng, Fan Zhang, David Bull
2022 arXiv   pre-print
The integrated codec has been submitted to the Challenge on Learned Image Compression (CLIC) 2022 (video track), and the team name associated with this submission is BVI_VC.  ...  This method has been integrated with the Versatile Video Coding Test Model (VTM) 15.2 to enhance the visual quality of the final reconstructed content.  ...  (UGC) dataset [13] .  ... 
arXiv:2205.09458v1 fatcat:fbe6lbp4wvbadj2dfbu7wniyiu

Subjective and Objective Quality Assessment of High Frame Rate Videos

Pavan C Madhusudana, Xiangxu Yu, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C Bovik
2021 IEEE Access  
The LIVE-YT-HFR database has been made available online for public use and evaluation purposes, with hopes that it will help advance research in this exciting video technology direction.  ...  Towards advancing progression in this direction we designed a new subjective resource, called the LIVE-YouTube-HFR (LIVE-YT-HFR) dataset, which is comprised of 480 videos having 6 different frame rates  ...  After graduation, he joined the Media Algorithm team in Youtube/Google. His research fields include video processing infrastructure, video quality assessment, and video compression.  ... 
doi:10.1109/access.2021.3100462 fatcat:evuzksl75ze2pgga3pf46cdw34
« Previous Showing results 1 — 15 out of 116 results