CaricatureGS: Exaggerating 3D Gaussian Splatting Faces With Gaussian Curvature

2026-01-06

arXiv.org

AI Summary

This work addresses the challenge of generating realistic and controllable 3D caricatured faces that balance geometric exaggeration, detail preservation, and real-time interactivity. The proposed method integrates Gaussian curvature–driven geometric deformation with a 3D Gaussian Splatting (3DGS) representation: exaggerated facial meshes are generated via a curvature-weighted Poisson equation, and pseudo-photorealistic caricature images are synthesized using local affine transformations. To unify natural and exaggerated expressions within a single Gaussian set, an alternating supervision training strategy is introduced, alongside an efficient surface interpolation scheme that enables continuous intensity control and localized editing. Experiments demonstrate that the approach outperforms existing methods in geometric controllability, visual realism, and real-time rendering performance.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Learning with Manifolds
• Humans and AI: Game Design — Procedural Content Generation & Storytelling

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications
• User Modeling, Personalization and Recommendation: User modeling and simulation for interactive and conversational systems

Abstract

A photorealistic and controllable 3D caricaturization framework for faces is introduced. We start with an intrinsic Gaussian curvature-based surface exaggeration technique, which, when coupled with texture, tends to produce over-smoothed renders. To address this, we resort to 3D Gaussian Splatting (3DGS), which has recently been shown to produce realistic free-viewpoint avatars. Given a multiview sequence, we extract a FLAME mesh, solve a curvature-weighted Poisson equation, and obtain its exaggerated form. However, directly deforming the Gaussians yields poor results, necessitating the synthesis of pseudo-ground-truth caricature images by warping each frame to its exaggerated 2D representation using local affine transformations. We then devise a training scheme that alternates real and synthesized supervision, enabling a single Gaussian collection to represent both natural and exaggerated avatars. This scheme improves fidelity, supports local edits, and allows continuous control over the intensity of the caricature. In order to achieve real-time deformations, an efficient interpolation between the original and exaggerated surfaces is introduced. We further analyze and show that it has a bounded deviation from closed-form solutions. In both quantitative and qualitative evaluations, our results outperform prior work, delivering photorealistic, geometry-controlled caricature avatars.

Problem

• 3D caricaturization
• Gaussian curvature
• 3D Gaussian Splatting
• face exaggeration
• realistic avatars

Innovation

• 3D Gaussian Splatting
• Gaussian curvature
• caricature generation
• real-time deformation
• geometry exaggeration