{"ID":2825669,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.20033","arxiv_id":"2512.20033","title":"FlashLips: 100-FPS Mask-Free Latent Lip-Sync using Reconstruction Instead of Diffusion or GANs","abstract":"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.","short_abstract":"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 edi...","url_abs":"https://arxiv.org/abs/2512.20033","url_pdf":"https://arxiv.org/pdf/2512.20033v3","authors":"[\"Andreas Zinonos\",\"Michał Stypułkowski\",\"Antoni Bigata\",\"Stavros Petridis\",\"Maja Pantic\",\"Nikita Drobyshev\"]","published":"2025-12-23T03:54:48Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\",\"Transformer\",\"Generative Adversarial Network\"]","has_code":false}