EasyVFX: Frequency-Driven Decoupling for Resource-Efficient VFX Generation

TL;DR

EasyVFX introduces a frequency-domain decoupling framework that significantly reduces resource requirements for high-quality visual effects generation by separating spatial and temporal components.

Contribution

The paper presents a novel frequency-aware mixture-of-experts architecture and a test-time training strategy that enable efficient and adaptable VFX synthesis with limited data and computational resources.

Findings

Achieves realistic VFX with fewer GPU resources.

Enables rapid adaptation to unseen effects with minimal steps.

Produces visually consistent and high-quality effects.

Abstract

Generating high-fidelity visual effects (VFX) typically demands massive datasets and prohibitive computational power due to the intricate coupling of spatial textures and temporal dynamics. In this paper, we introduce EasyVFX, a resource-efficient framework that achieves realistic VFX synthesis under stringent constraints. Our core philosophy lies in frequency-domain decomposition: we observe that the complexity of VFX can be significantly mitigated by decoupling high-frequency components, which represent intricate spatial appearances, from low-frequency components that encapsulate global motion dynamics. This spectral disentanglement transforms a high-dimensional learning problem into manageable sub-tasks, thereby lowering the optimization barrier and reducing data dependency. Building upon this insight, we propose a two-stage training paradigm. First, we design a Frequency-aware Mixture-of-Experts (Freq-MoE) architecture. By utilizing a soft routing mechanism, our model assigns specialized experts to distinct spectral bands, enabling them to cultivate robust priors for appearance and motion dynamics. This specialization allows the model to acquire foundational VFX knowledge with fewer GPU resources. Second, we introduce a Test-Time Training strategy powered by a novel Frequency-constraint Loss. This allows the pre-trained model to swiftly adapt to specific, unseen effects through localized optimizations, requiring only about 100 steps on a single GPU. Experimental results demonstrate that EasyVFX produces structurally consistent and visually stunning effects, proving that frequency-aware learning is a key catalyst for democratizing professional-grade VFX.