Paper · arXiv 2412.07583
Mobile Video Diffusion
Haitam Ben Yahia, Denis Korzhenkov, Ioannis Lelekas, Amir Ghodrati, Amirhossein Habibian
A mobile-optimized video diffusion model reduces computational demands and maintains quality through pruning and adversarial finetuning.
Abstract
Video diffusion models have achieved impressive realism and controllability but are limited by high computational demands, restricting their use on mobile devices. This paper introduces the first mobile-optimized video diffusion model. Starting from a spatio-temporal UNet from Stable Video Diffusion (SVD), we reduce memory and computational cost by reducing the frame resolution, incorporating multi-scale temporal representations, and introducing two novel pruning schema to reduce the number of channels and temporal blocks. Furthermore, we employ adversarial finetuning to reduce the denoising to a single step. Our model, coined as MobileVD, is 523x more efficient (1817.2 vs. 4.34 TFLOPs) with a slight quality drop (FVD 149 vs. 171), generating latents for a 14x512x256 px clip in 1.7 seconds on a Xiaomi-14 Pro. Our results are available at https://qualcomm-ai-research.github.io/mobile-video-diffusion/