DiffVC-ONE: Diffusion-based Generative Video Compression with One-Step Video Diffusion Transformer

DiffVC-ONE：基于一步视频扩散 Transformer 的生成式视频压缩

Generative video compression can recover rich visual details at low bitrates, but simultaneously achieving high temporal consistency and low inference cost remains challenging. 生成式视频压缩能够在低比特率下恢复丰富的视觉细节，但同时实现高时间一致性和低推理成本仍然具有挑战性。

To address this issue, we propose DiffVC-ONE, a diffusion-based generative video compression framework built on a one-step Video Diffusion Transformer. 为了解决这一问题，我们提出了 DiffVC-ONE，这是一个基于一步视频扩散 Transformer（Video Diffusion Transformer）构建的生成式视频压缩框架。

First, we introduce a Unified Unidirectional Latent Compressor that uses a shared model to efficiently and uniformly compress compact latent slices. 首先，我们引入了一种统一的单向潜在压缩器（Unified Unidirectional Latent Compressor），它使用共享模型来高效且统一地压缩紧凑的潜在切片。

We then develop a Video DiT-based One-Step Diffusion Enhancer that uses the reconstructed latent slices as content anchors and performs single-step spatio-temporal perceptual enhancement over an entire group of pictures. 随后，我们开发了一种基于视频 DiT 的一步扩散增强器（Video DiT-based One-Step Diffusion Enhancer），它利用重构的潜在切片作为内容锚点，并对整个图像组（GOP）执行单步时空感知增强。

Finally, a Hybrid Condition Generator extracts structural, strength, and semantic conditions from the reconstructed content and quantization information. 最后，混合条件生成器（Hybrid Condition Generator）从重构内容和量化信息中提取结构、强度和语义条件。

These conditions preserve faithful regions, control the degree of generative enhancement, and supplement content-aware perceptual details during one-step diffusion enhancement. 这些条件在一步扩散增强过程中，能够保留真实区域、控制生成增强的程度，并补充内容感知的感知细节。

Extensive experiments on multiple standard benchmarks demonstrate that DiffVC-ONE achieves state-of-the-art perceptual quality and temporal consistency with low inference cost. 在多个标准基准测试上的广泛实验表明，DiffVC-ONE 以较低的推理成本实现了最先进的感知质量和时间一致性。