A Tri-Dynamic Preprocessing Framework for UGC Video Compression

TL;DR

This paper introduces a Tri-Dynamic Preprocessing framework for UGC video compression that adapts preprocessing, quantization, and rate-distortion tradeoff to handle the high variability of user-generated videos, improving compression performance.

Contribution

The paper presents a novel adaptive preprocessing framework specifically designed for the diverse and variable nature of UGC videos, enhancing compression effectiveness.

Findings

Achieves superior compression performance on large-scale UGC datasets.

Effectively adapts to diverse UGC video characteristics.

Outperforms existing methods in rate-distortion efficiency.

Abstract

In recent years, user generated content (UGC) has become the dominant force in internet traffic. However, UGC videos exhibit a higher degree of variability and diverse characteristics compared to traditional encoding test videos. This variance challenges the effectiveness of data-driven machine learning algorithms for optimizing encoding in the broader context of UGC scenarios. To address this issue, we propose a Tri-Dynamic Preprocessing framework for UGC. Firstly, we employ an adaptive factor to regulate preprocessing intensity. Secondly, an adaptive quantization level is employed to fine-tune the codec simulator. Thirdly, we utilize an adaptive lambda tradeoff to adjust the rate-distortion loss function. Experimental results on large-scale test sets demonstrate that our method attains exceptional performance.