FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zinonos, Andreas, Stypułkowski, Michał, Bigata, Antoni, Petridis, Stavros, Pantic, Maja, Drobyshev, Nikita
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915944115208192
author Zinonos, Andreas
Stypułkowski, Michał
Bigata, Antoni
Petridis, Stavros
Pantic, Maja
Drobyshev, Nikita
author_facet Zinonos, Andreas
Stypułkowski, Michał
Bigata, Antoni
Petridis, Stavros
Pantic, Maja
Drobyshev, Nikita
contents We present FlashLips, a two-stage, mask-free lip-sync system that decouples lips control from rendering and achieves real-time performance, with our U-Net variant running at over 100 FPS on a single GPU, while matching the visual quality of larger state-of-the-art models. Stage 1 is a compact, one-step latent-space editor that reconstructs an image using a reference identity, a masked target frame, and a low-dimensional lips-pose vector, trained purely with reconstruction losses - no GANs or diffusion. To remove explicit masks at inference, we use self-supervision via mouth-altered target variants as pseudo ground truth, teaching the network to localize lip edits while preserving the rest. Stage 2 is an audio-to-pose transformer trained with a flow-matching objective to predict lips-pose vectors from speech. Together, these stages form a simple and stable pipeline that combines deterministic reconstruction with robust audio control, delivering high perceptual quality and faster-than-real-time speed.
format Preprint
id arxiv_https___arxiv_org_abs_2512_20033
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs
Zinonos, Andreas
Stypułkowski, Michał
Bigata, Antoni
Petridis, Stavros
Pantic, Maja
Drobyshev, Nikita
Computer Vision and Pattern Recognition
We present FlashLips, a two-stage, mask-free lip-sync system that decouples lips control from rendering and achieves real-time performance, with our U-Net variant running at over 100 FPS on a single GPU, while matching the visual quality of larger state-of-the-art models. Stage 1 is a compact, one-step latent-space editor that reconstructs an image using a reference identity, a masked target frame, and a low-dimensional lips-pose vector, trained purely with reconstruction losses - no GANs or diffusion. To remove explicit masks at inference, we use self-supervision via mouth-altered target variants as pseudo ground truth, teaching the network to localize lip edits while preserving the rest. Stage 2 is an audio-to-pose transformer trained with a flow-matching objective to predict lips-pose vectors from speech. Together, these stages form a simple and stable pipeline that combines deterministic reconstruction with robust audio control, delivering high perceptual quality and faster-than-real-time speed.
title FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.20033