arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2107.13136eess.IVcs.CVcs.LG
收藏

来自生成建模的神经视频压缩洞见

Insights from Generative Modeling for Neural Video Compression

University of California Irvine(加州大学欧文分校);DeepMind

Ruihan Yang, Yibo Yang, Joseph Marino, Stephan Mandt

AI总结:

本文以深度自回归和潜变量生成模型统一审视神经视频编解码器,提出改进的时间变换、结构化熵模型和可变比特率架构,实现高分辨率视频压缩的最优性能。

摘要:

尽管近期机器学习研究已经揭示了 VAE 等深度生成模型与学习式压缩中使用的率失真损失之间的联系,但这类工作大多聚焦于图像。本着类似精神,我们通过深度自回归建模和潜变量建模的视角来审视近期提出的神经视频编码算法。我们将这些编解码器表示为广义随机时间自回归变换的实例,并受归一化流和结构化先验启发,提出进一步改进的新途径。我们提出若干架构,在高分辨率视频上取得最先进的视频压缩性能,并讨论它们的权衡与消融实验。具体而言,我们提出:(i)改进的时间自回归变换;(ii)具有结构化和时间依赖关系的改进熵模型;以及(iii)我们算法的可变比特率版本。由于我们的改进与一大类现有模型兼容,我们进一步证明生成建模视角能够推动神经视频编码领域的发展。

英文摘要:

While recent machine learning research has revealed connections between deep generative models such as VAEs and rate-distortion losses used in learned compression, most of this work has focused on images. In a similar spirit, we view recently proposed neural video coding algorithms through the lens of deep autoregressive and latent variable modeling. We present these codecs as instances of a generalized stochastic temporal autoregressive transform, and propose new avenues for further improvements inspired by normalizing flows and structured priors. We propose several architectures that yield state-of-the-art video compression performance on high-resolution video and discuss their tradeoffs and ablations. In particular, we propose (i) improved temporal autoregressive transforms, (ii) improved entropy models with structured and temporal dependencies, and (iii) variable bitrate versions of our algorithms. Since our improvements are compatible with a large class of existing models, we provide further evidence that the generative modeling viewpoint can advance the neural video coding field.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑