Video transcoding optimization based on input perceptual quality

Wang, Yilin; Talebi, Hossein; Yang, Feng; Yim, Joong Gon; Birkbeck, Neil; Adsumilli, Balu; Milanfar, Peyman

doi:10.1117/12.2569332

Cited by 8 publications

(7 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The binarizing threshold T is automatically determined by the GMM clustering described above. This binary categorization problem is particularly interesting since it caters to applications involving optimizing transcoding configurations for low and high input quality separately, such as the QGT framework [26]. An alternative way of achieving binary predictions is to fit a regressor to the MOS labels, then apply a threshold to obtain binary predictions.…”

Section: Task B: Binary Classificationmentioning

confidence: 99%

“…Like the IAA problem [21], relaxing regression to (binary) classification could hence make this problem more tractable. Second, inspired by just-noticeabledifference (JND) [24] approaches to reference-based video quality, we suggest that for blind visual quality prediction, similar JND-like approaches may be taken to exploit the visual discriminative power and limits of subjects' quality per- [26], which involves encoding videos uploaded to YouTube with parameters optimized based on its input quality category: {low, medium, high}.…”

Section: Introductionmentioning

confidence: 99%

“…Finally, such an approach may be more useful on UGC media transcoding platforms like YouTube and Facebook, i.e., classifying quality scores onto categories, since only discrete quality-guided decisions can be deployed when (pre-)processing UGC content. One example that helps explain this is the quality-guided transcoding (QGT) framework proposed in [26], which involves encoding videos uploaded to YouTube with parameters optimized based on its input quality category: {low, medium, high}.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Regression or classification? New methods to evaluate no-reference picture and video quality models

Chen

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

Video and image quality assessment has long been projected as a regression problem, which requires predicting a continuous quality score given an input stimulus. However, recent efforts have shown that accurate quality score regression on real-world user-generated content (UGC) is a very challenging task. To make the problem more tractable, we propose two new methods -binary, and ordinal classification -as alternatives to evaluate and compare no-reference quality models at coarser levels. Moreover, the proposed new tasks convey more practical meaning on perceptually optimized UGC transcoding, or for preprocessing on media processing platforms. We conduct a comprehensive benchmark experiment of popular no-reference quality models on recent in-the-wild picture and video quality datasets, providing reliable baselines for both evaluation methods to support further studies. We hope this work promotes coarse-grained perceptual modeling and its applications to efficient UGC processing.

show abstract

Section: Task B: Binary Classificationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Regression or classification? New methods to evaluate no-reference picture and video quality models

Chen

et al. 2021

ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)

Self Cite

View full text Add to dashboard Cite

show abstract

“…Y. Wang, N. Birkbeck, and B. Adsumilli are with YouTube Media Algorithms Team, Google LLC, Mountain View, CA, 94043, USA. (emails: yilin@google.com, birkbeck@google.com, badsumilli@google.com) to be able to deploy flexible video transcoding profiles in industry-level applications based on measurements of input video quality to achieve even better rate-quality tradeoffs relative to traditional encoding paradigms [5]. The decision tuning strategy of such an adaptive encoding scheme, however, would require the guidance of an accurate and efficient NR or blind video quality (BVQA) model suitable for UGC [6].…”

Section: Introductionmentioning

confidence: 99%

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

Wang

et al. 2021

IEEE Open J. Signal Process.

Self Cite

131

View full text Add to dashboard Cite

Blind or no-reference video quality assessment of user-generated content (UGC) has become a trending, challenging, heretofore unsolved problem. 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. Previous studies have shown that natural scene statistics and deep learning features are both sufficient to capture spatial distortions, which contribute to a significant aspect of UGC video quality issues. However, these models are either incapable or inefficient for predicting the quality of complex and diverse UGC videos in practical applications. Here we introduce an effective and efficient video quality model for UGC content, which we dub the Rapid and Accurate Video Quality Evaluator (RAPIQUE), which we show performs comparably to state-of-the-art (SOTA) models but with orders-of-magnitude faster runtime. RAPIQUE combines and leverages the advantages of both quality-aware scene statistics features and semantics-aware deep convolutional features, allowing us to design the first general and efficient spatial and temporal (space-time) bandpass statistics model for video quality modeling. 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. We hope this work promotes and inspires further efforts towards practical modeling of video quality problems for potential real-time and low-latency applications. To promote public usage, an implementation of RAPIQUE has been made freely available online: https://github.com/vztu/RAPIQUE.

show abstract

“…One intriguing property of UGC videos, from the data compression aspect, is that they often suffer from artifacts or distortions in the originals. Accordingly, one could ostensibly deploy flexible transcoding profiles based on measurements of the input quality to achieve better rate-quality tradeoffs [1]. The decision tuning strategy of such an adaptive encoding scheme, however, would require the guidance of an accurate and efficient no-reference (NR) or blind video quality (BVQA) model suitable for UGC [2].…”

Section: Introductionmentioning

confidence: 99%

A Temporal Statistics Model For UGC Video Quality Prediction

Chen

Wang

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

Blind video quality assessment of user-generated content (UGC) has become a trending and challenging problem. Previous studies have shown the efficacy of natural scene statistics for capturing spatial distortions. The exploration of temporal video statistics on UGC, however, is relatively limited. Here we propose the first general, effective and efficient temporal statistics model accounting for temporal-or motion-related distortions for UGC video quality assessment, by analyzing regularities in the temporal bandpass domain. The proposed temporal model can serve as a plug-in module to boost existing no-reference video quality predictors that lack motion-relevant features. Our experimental results on recent large-scale UGC video databases show that the proposed model can significantly improve the performances of existing methods, at a very reasonable computational expense.

show abstract

Video transcoding optimization based on input perceptual quality

Cited by 8 publications

References 18 publications

Regression or classification? New methods to evaluate no-reference picture and video quality models

Regression or classification? New methods to evaluate no-reference picture and video quality models

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

A Temporal Statistics Model For UGC Video Quality Prediction

Contact Info

Product

Resources

About