Avoiding Quality Saturation in UGC Compression Using Denoised References

2025-11-20

AI Summary

A prevalent quality saturation (QS) problem in user-generated content (UGC) video compression—where increasing bitrate fails to improve visual quality and instead entrenches original artifacts—is addressed in this work. We propose a novel compression optimization framework leveraging denoised reference videos: for the first time, a denoised video serves as the distortion reference to define a transform-domain, block-level D-MSE metric; based on this, we formally characterize QS and design lightweight preprocessing modules—DSD and RDSD—that identify and bypass QS-prone regions without iterative encoding–decoding. The method integrates noise-source coding theory, no-reference quality assessment, and low-complexity Lagrangian parameter estimation. Evaluated on AVC, our approach effectively avoids the QS regime, achieving 8%–20% BD-rate savings across multiple no-reference quality metrics.

Technology Category

• Search and Optimization: Distributed Search
• Data Mining & Knowledge Management: Data Compression
• Computer Vision: Learning & Optimization for CV

Application Category

• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Web Mining and Content Analysis: Web data quality in the era of algorithmically-generated content

Abstract

Video-sharing platforms must re-encode large volumes of noisy user-generated content (UGC) to meet streaming demands. However, conventional codecs, which aim to minimize the mean squared error (MSE) between the compressed and input videos, can cause quality saturation (QS) when applied to UGC, i.e., increasing the bitrate preserves input artifacts without improving visual quality. A direct approach to solve this problem is to detect QS by repeatedly evaluating a non-reference metric (NRM) on videos compressed with multiple codec parameters, which is inefficient. In this paper, we re-frame UGC compression and QS detection from the lens of noisy source coding theory: rather than using a NRM, we compute the MSE with respect to the denoised UGC, which serves as an alternative reference (D-MSE). Unlike MSE measured between the UGC input and the compressed UGC, D-MSE saturates at non-zero values as bitrates increase, a phenomenon we term distortion saturation (DS). Since D-MSE can be computed at the block level in the transform domain, we can efficiently detect D-MSE without coding and decoding with various parameters. We propose two methods for DS detection: distortion saturation detection (DSD), which relies on an input-dependent threshold derived from the D-MSE of the input UGC, and rate-distortion saturation detection (RDSD), which estimates the Lagrangian at the saturation point using a low-complexity compression method. Both methods work as a pre-processing step that can help standard-compliant codecs avoid QS in UGC compression. Experiments with AVC show that preventing encoding in the saturation region, i.e., avoiding encoding at QPs that result in QS according to our methods, achieves BD-rate savings of 8%-20% across multiple different NRMs, compared to a naïve baseline that encodes at the given input QP while ignoring QS.

Problem

• Preventing quality saturation in UGC compression using denoised references
• Detecting distortion saturation efficiently without repeated coding evaluations
• Avoiding unnecessary bitrate increases that preserve input artifacts

Innovation

• Using denoised UGC as alternative reference
• Detecting distortion saturation at block level
• Proposing two methods to avoid quality saturation