This repository contains the code for a GitHub pages site for the publication MoViE: Mobile Diffusion for Video Editing. The project page is hosted at https://qualcomm-ai-research.github.io/mobile-video-editing/.

MoViE is a high-throughput, zero-shot video diffusion pipeline designed for mobile devices. MoViE enables video editing at 12 frames per second on mobile devices.

Recent progress in diffusion-based video editing has shown remarkable potential for practical applications. However, these methods remain prohibitively expensive and challenging to deploy on mobile devices. In this study, we introduce a series of optimizations that render mobile video editing feasible. Building upon the existing image editing model, we first optimize its architecture and incorporate a lightweight autoencoder. Subsequently, we extend classifier-free guidance distillation to multiple modalities, resulting in a threefold on-device speedup. Finally, we reduce the number of sampling steps to one by introducing a novel adversarial distillation scheme which preserves the controllability of the editing process. Collectively, these optimizations enable video editing at 12 frames per second on mobile devices, while maintaining high quality.