For neural video codec, it is critical, yet challenging, to design an
efficient entropy model which can accurately predict the probability
distribution of the quantized latent representation. However, most existing
video codecs directly use the ready-made entropy model from image codec to
encode the residual or motion, and do not fully leverage the spatial-temporal
characteristics in video. To this end, this paper proposes a powerful entropy
model which efficiently captures both spatial and temporal dependencies. In
particular, we introduce the latent prior which exploits the correlation among
the latent representation to squeeze the temporal redundancy. Meanwhile, the
dual spatial prior is proposed to reduce the spatial redundancy in a
parallel-friendly manner. In addition, our entropy model is also versatile.
Besides estimating the probability distribution, our entropy model also
generates the quantization step at spatial-channel-wise. This content-adaptive
quantization mechanism not only helps our codec achieve the smooth rate
adjustment in single model but also improves the final rate-distortion
performance by dynamic bit allocation. Experimental results show that, powered
by the proposed entropy model, our neural codec can achieve 18.2% bitrate
saving on UVG dataset when compared with H.266 (VTM) using the highest
compression ratio configuration. It makes a new milestone in the development of
neural video codec. The codes are at this https URL

本文提出了一种强大的熵模型，能够高效地捕捉视频中的空间和时间依赖关系，使用潜在先验来减少时间冗余，使用双重空间先验来并行地减少空间冗余。此外，该熵模型还具有内容自适应量化机制，有助于编解码器实现平滑的速率调整，并通过动态位分配改善最终的速率失真性能。实验结果表明，使用该熵模型作为支撑，与最高压缩比配置下的 H.266（VTM）相比，我们的神经编解码器可以在 UVG 数据集上实现 18.2% 的比特率节省，这是神经视频编解码器发展的新里程碑。

混合时空熵建模用于神经网络视频压缩

Hybrid Spatial-Temporal Entropy Modelling for Neural Video Compression

A new algorithmic framework is proposed for learning autoencoders of data
distributions. We minimize the discrepancy between the model and target
distributions, with a \emph{relational regularization} on the learnable latent
prior. This regularization penalizes the fused Gromov-Wasserstein (FGW)
distance between the latent prior and its corresponding posterior, allowing one
to flexibly learn a structured prior distribution associated with the
generative model. Moreover, it helps co-training of multiple autoencoders even
if they have heterogeneous architectures and incomparable latent spaces. We
implement the framework with two scalable algorithms, making it applicable for
both probabilistic and deterministic autoencoders. Our relational regularized
autoencoder (RAE) outperforms existing methods, $e.g.$, the variational
autoencoder, Wasserstein autoencoder, and their variants, on generating images.
Additionally, our relational co-training strategy for autoencoders achieves
encouraging results in both synthesis and real-world multi-view learning tasks.
The code is at this https URL Relational-AutoEncoders.

提出了一种新的算法框架用于学习数据分布的自编码器，该框架最小化模型与目标分布的差异，通过对可学习潜在先验的关系正则化，惩罚潜在先验与其对应后验之间的融合 Gromov-Wasserstein（FGW）距离，从而允许灵活地学习与生成模型相关的结构化先验分布，帮助多个自编码器的联合训练，即使它们具有异构的体系结构和无法比较的潜在空间，通过两种可扩展的算法实现该框架，适用于概率和确定性自编码器，我们的关系正则化自编码器（RAE）在生成图像方面优于现有方法，例如变分自编码器，Wasserstein 自编码器及其变体，此外，我们的关系联合训练策略也取得了在合成和真实世界多视图学习任务中鼓舞人心的结果。