Versatile Video Tokenization with Generative 2D Gaussian Splatting

TL;DR

The paper introduces GVT, a versatile video tokenizer using generative 2D Gaussian Splatting that adaptively encodes spatial and temporal content, improving video reconstruction, recognition, and compression.

Contribution

It proposes a novel Gaussian Video Transformer with generative 2D Gaussians and strategies for spatial adaptability and temporal separation, enhancing versatility over fixed-grid methods.

Findings

Achieves state-of-the-art video reconstruction quality.

Outperforms baseline in action recognition tasks.

Provides comparable compression performance.

Abstract

Video tokenization procedure is critical for a wide range of video processing tasks. Most existing approaches directly transform video into fixed-grid and patch-wise tokens, which exhibit limited versatility. Spatially, uniformly allocating a fixed number of tokens often leads to over-encoding in low-information regions. Temporally, reducing redundancy remains challenging without explicitly distinguishing between static and dynamic content. In this work, we propose the Gaussian Video Transformer (GVT), a versatile video tokenizer built upon a generative 2D Gaussian Splatting (2DGS) strategy. We first extract latent rigid features from a video clip and represent them with a set of 2D Gaussians generated by our proposed Spatio-Temporal Gaussian Embedding (STGE) mechanism in a feed-forward manner. Such generative 2D Gaussians not only enhance spatial adaptability by assigning higher (resp., lower) rendering weights to regions with higher (resp., lower) information content during rasterization, but also improve generalization by avoiding per-video optimization.To enhance the temporal versatility, we introduce a Gaussian Set Partitioning (GSP) strategy that separates the 2D Gaussians into static and dynamic sets, which explicitly model static content shared across different time-steps and dynamic content specific to each time-step, enabling a compact representation.We primarily evaluate GVT on the video reconstruction, while also assessing its performance on action recognition and compression using the UCF101, Kinetics, and DAVIS datasets. Extensive experiments demonstrate that GVT achieves a state-of-the-art video reconstruction quality, outperforms the baseline MAGVIT-v2 in action recognition, and delivers comparable compression performance.