Online free-view navigation in volumetric videos requires high-quality rendering and real-time streaming in order to provide immersive user experiences. However, existing methods may not handle dynamic scenes with complex motions, and their models (e.g., dynamic NeRF and 3DGS) may not be streamable due to storage and bandwidth constraints. In this paper, we propose a novel 4D Gaussian Video (4DGV) approach that enables the creation and streaming of photorealistic, volumetric videos for dynamic scenes over the Internet. The core of our 4DGV is a novel streamable group of Gaussians (GOG) representation based on motion layering. Each GOG consists of static and dynamic points obtained via lifting 2D segmentation into 3D in motion layering, where the deformation of each dynamic point is represented as the temporal offset of its attributes. We also adaptively convert static points back to dynamic points to handle the appearance change, (e.g. moving shadows and reflections), of static objects through optimization. To support real-time streaming of 4DGV, we show that by applying quantization on Gaussian attributes and H.265 encoding on deformation offsets, our GOG representation can be significantly compressed (to around 6% of the original model size) without sacrificing the accuracy (PSNR loss less than 0.01dB). Extensive experiments on standard benchmarks demonstrate that our method outperforms state-of-the-art volumetric video approaches, with superior rendering quality and minimum storage overheads.

Real-time free-viewpoint rendering of reconstructed 4DGV (videos downsized).

Adaptive dynamic/static point separation.

Video comparisons on public datasets (more results here).