Style4D-Bench: A Benchmark Suite for 4D Stylization

2025-08-26

AI Summary

The lack of a standardized evaluation benchmark for 4D stylization hinders systematic progress in this emerging field. To address this, we introduce Style4D—the first dedicated benchmark suite for 4D stylization, comprising a rigorous evaluation protocol, a high-quality dynamic scene dataset, and strong baseline methods. Methodologically, we build upon 4D Gaussian Splatting for spatiotemporal scene representation and propose a lightweight Gaussian MLP alongside a geometry-preserving global style transfer module. Our framework integrates contrastive consistency learning with structural and content constraints to achieve multi-view consistent, temporally stable, and detail-faithful stylized rendering. Evaluated on our curated dataset, our approach achieves state-of-the-art performance, significantly improving rendering quality, inter-frame and inter-view consistency, and quantitative assessability. Style4D establishes the first standardized foundation for reproducible research and fair comparison in 4D stylization.

Technology Category

• Computer Vision: 3D Computer Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Natural Language Processing: Sentiment Analysis, Stylistic Analysis, and Argument Mining

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

We introduce Style4D-Bench, the first benchmark suite specifically designed for 4D stylization, with the goal of standardizing evaluation and facilitating progress in this emerging area. Style4D-Bench comprises: 1) a comprehensive evaluation protocol measuring spatial fidelity, temporal coherence, and multi-view consistency through both perceptual and quantitative metrics, 2) a strong baseline that make an initial attempt for 4D stylization, and 3) a curated collection of high-resolution dynamic 4D scenes with diverse motions and complex backgrounds. To establish a strong baseline, we present Style4D, a novel framework built upon 4D Gaussian Splatting. It consists of three key components: a basic 4DGS scene representation to capture reliable geometry, a Style Gaussian Representation that leverages lightweight per-Gaussian MLPs for temporally and spatially aware appearance control, and a Holistic Geometry-Preserved Style Transfer module designed to enhance spatio-temporal consistency via contrastive coherence learning and structural content preservation. Extensive experiments on Style4D-Bench demonstrate that Style4D achieves state-of-the-art performance in 4D stylization, producing fine-grained stylistic details with stable temporal dynamics and consistent multi-view rendering. We expect Style4D-Bench to become a valuable resource for benchmarking and advancing research in stylized rendering of dynamic 3D scenes. Project page: https://becky-catherine.github.io/Style4D . Code: https://github.com/Becky-catherine/Style4D-Bench .

Problem

• Establishing first benchmark for 4D stylization evaluation
• Developing framework for spatio-temporal consistent 4D stylization
• Creating standardized metrics for dynamic 3D style transfer

Innovation

• 4D Gaussian Splatting scene representation
• Style Gaussian Representation with MLPs
• Geometry-Preserved Style Transfer module