Splat4D: Diffusion-Enhanced 4D Gaussian Splatting for Temporally and Spatially Consistent Content Creation

Generating high-quality 4D content from monocular videos—for applications such as digital humans and AR/VR—poses challenges in ensuring temporal and spatial consistency, preserving intricate details, and incorporating user guidance effectively. To overcome these challenges, we introduce Splat4D, a novel framework enabling high-fidelity 4D content generation from a monocular video. Splat4D achieves superior performance while maintaining faithful spatial-temporal coherence, by leveraging multi-view rendering, inconsistency identification, a video diffusion model, and an asymmetric U-Net for refinement. Through extensive evaluations on public benchmarks, Splat4D consistently demonstrates state-of-the-art performance across various metrics, underscoring the efficacy of our approach. Additionally, the versatility of Splat4D is validated in various applications such as text/image conditioned 4D generation, 4D human generation, and text-guided content editing, producing coherent outcomes following user instructions

Our method for 4D content generation begins with processing input data (text, image, or monocular video) to produce high-quality multi-view image sequences. These sequences are used to initialize a 4D Gaussian representation via an asymmetry U-Net and image splattering. Refinement steps include leveraging uncertainty masking and video denoising diffusion to ensure high fidelity and spatial-temporal consistency, culminating in versatile 4D content creation. The pipeline supports optional text-guided content editing, enabling dynamic modifications of the 4D output for enhanced flexibility and creative control.