ActionMesh: Animated 3D Mesh Generation with Temporal 3D Diffusion

2026-01-22

AI Summary

Existing approaches to animating 3D meshes are often hindered by complex setups, low efficiency, or insufficient output quality. This work proposes ActionMesh, the first method to incorporate temporal dynamics into 3D diffusion models by introducing a temporally coherent 3D diffusion framework. Leveraging a temporal 3D autoencoder, ActionMesh enables end-to-end generation of topologically consistent, high-fidelity animated 3D meshes from multimodal inputs—including monocular videos, text prompts, or static meshes paired with text—without requiring skeletal rigging. Evaluated on the Consistent4D and Objaverse benchmarks, the method achieves state-of-the-art performance in both geometric accuracy and temporal consistency, significantly improving generation speed and visual quality. Furthermore, it naturally supports downstream applications such as texture mapping and motion retargeting.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Deep Generative Models & Autoencoders
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Responsible Web: Machine-in-the-loop, human agency and autonomy
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Generating animated 3D objects is at the heart of many applications, yet most advanced works are typically difficult to apply in practice because of their limited setup, their long runtime, or their limited quality. We introduce ActionMesh, a generative model that predicts production-ready 3D meshes"in action"in a feed-forward manner. Drawing inspiration from early video models, our key insight is to modify existing 3D diffusion models to include a temporal axis, resulting in a framework we dubbed"temporal 3D diffusion". Specifically, we first adapt the 3D diffusion stage to generate a sequence of synchronized latents representing time-varying and independent 3D shapes. Second, we design a temporal 3D autoencoder that translates a sequence of independent shapes into the corresponding deformations of a pre-defined reference shape, allowing us to build an animation. Combining these two components, ActionMesh generates animated 3D meshes from different inputs like a monocular video, a text description, or even a 3D mesh with a text prompt describing its animation. Besides, compared to previous approaches, our method is fast and produces results that are rig-free and topology consistent, hence enabling rapid iteration and seamless applications like texturing and retargeting. We evaluate our model on standard video-to-4D benchmarks (Consistent4D, Objaverse) and report state-of-the-art performances on both geometric accuracy and temporal consistency, demonstrating that our model can deliver animated 3D meshes with unprecedented speed and quality.

Problem

• animated 3D mesh generation
• temporal consistency
• 3D diffusion
• production-ready 3D assets
• 4D generation

Innovation

• temporal 3D diffusion
• animated 3D mesh generation
• rig-free animation
• topology-consistent deformation
• feed-forward 4D generation