Recent advancements in video compression are focusing on enhancing perceptual quality and efficiency, particularly at ultra-low bitrates, which is crucial for applications like streaming and video conferencing. Techniques such as Efficient Video Diffusion with Sparse Information Transmission are addressing the limitations of traditional methods by improving temporal coherence and reducing blurriness in reconstructed frames. Meanwhile, frameworks like Generative Video Codec are pioneering zero-shot coding, allowing for direct bitstream transmission without retraining, thereby streamlining the compression process. Innovations like LRConv-NeRV are optimizing neural video compression by reducing computational demands while maintaining quality, making them suitable for resource-constrained environments. Additionally, methods such as ProGVC are integrating progressive transmission with efficient entropy coding, enabling flexible bitrate control and enhancing scalability. Collectively, these developments indicate a shift towards more sophisticated, adaptable compression techniques that prioritize both quality and efficiency, addressing the growing demand for high-performance video solutions in commercial settings.
Video compression aims to maximize reconstruction quality with minimal bitrates. Beyond standard distortion metrics, perceptual quality and temporal consistency are also critical. However, at ultra-lo...
Existing generative video compression methods use generative models only as post-hoc reconstruction modules atop conventional codecs. We propose \emph{Generative Video Codec} (GVC), a zero-shot framew...
Neural Representations for Videos (NeRV) encode entire video sequences within neural network parameters, offering an alternative paradigm to conventional video codecs. However, the convolutional decod...
Although learned video compression methods have exhibited outstanding performance, most of them typically follow a hybrid coding paradigm that requires explicit motion estimation and compensation, res...
Perceptual video compression leverages generative priors to reconstruct realistic textures and motions at low bitrates. However, existing perceptual codecs often lack native support for variable bitra...