cs.CVNov 9, 2026

MPEG-Diffusion: Mixed-Resolution Video Generation

Authors: Dennis Menn, Tianpei Gu, Guande He, Aryan V S, Yanke Song, Nan Liu, Desai Xie, Xun Huang, 

Organizations: UT Austin · Roblox

Abstract

Video diffusion transformers (DiTs) generate realistic videos, but their heavy computational cost limits broader real-world applications. Mixed-resolution generation reduces this cost by generating the visually important region at high resolution (HR) and the rest at low resolution (LR), yet existing methods suffer a severe quality drop on videos and require knowing the HR region in advance. We show that these failures are not inherent to mixed resolution, and propose MPEG-Diffusion\textbf{MPEG-Diffusion} to address their three root causes: (1) dual-resolution attention to mitigate the curse of mixed resolution, where each query reads the K/V of its own resolution, (2) a mixed-resolution decoder to blend the HR and LR latents before decoding and resolve distortion, and (3) a self-selected HR region using the DiT's knowledge to find the visually important region without requiring it in advance. Processing only 30%30\% of full-resolution tokens, MPEG-Diffusion speeds up denoising by 4.8×4.8\times on Wan2.1 14B (3.8×3.8\times on 1.3B), outperforming prior methods on nearly every metric. In the user study, we demonstrate the high quality of MPEG-Diffusion, with raters preferring it over prior mixed-resolution methods in over 90%90\% of pairs and rating it close to full resolution.

Figures & tables

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. DraftAttention2: Fast Video Diffusion with Low-Resolution-Guided Mixed-Precision Attention

    Sep 26, 2026Rui Ding, Haopeng Li, Weize Ma +10Video Diffusion ModelsDiffusion Model Inference Acceleration

  2. Where Compute Matters: Heterogeneous Attention for Efficient Video Diffusion

    Sep 25, 2026Olga Zatsarynna, Denis Korzhenkov, Juergen Gall +2Video Diffusion ModelsDiffusion Model Inference Acceleration