Turbo4DGen: Ultra-Fast Acceleration for 4D Generation

TL;DR

Turbo4DGen is a novel framework that significantly accelerates 4D content generation by reusing attention computations and pruning, achieving nearly tenfold speedup without quality loss.

Contribution

It introduces a spatiotemporal cache, semantic-aware attention pruning, and an adaptive scheduler, pioneering dedicated acceleration for 4D generation.

Findings

Achieves 9.7× speedup on benchmark datasets.

Maintains quality comparable to non-accelerated methods.

First framework specifically designed for 4D generation acceleration.

Abstract

4D generation, or dynamic 3D content generation, integrates spatial, temporal, and view dimensions to model realistic dynamic scenes, playing a foundational role in advancing world models and physical AI. However, maintaining long-chain consistency across both frames and viewpoints through the unique spatio-camera-motion (SCM) attention mechanism introduces substantial computational and memory overhead, often leading to out-of-memory (OOM) failures and prohibitive generation times. To address these challenges, we propose Turbo4DGen, an ultra-fast acceleration framework for diffusion-based multi-view 4D content generation. Turbo4DGen introduces a spatiotemporal cache mechanism that persistently reuses intermediate attention across denoising steps, combined with dynamically semantic-aware attention pruning and an adaptive SCM chain bypass scheduler, to drastically reduce redundant SCM attention computation. Our experimental results show that Turbo4DGen achieves an average 9.7$\times$ speedup without quality degradation on the ObjaverseDy and Consistent4D datasets. To the best of our knowledge, Turbo4DGen is the first dedicated acceleration framework for 4D generation.