Skip to content

Source paper

FlashPortrait: 6x Faster Infinite Portrait Animation with Adaptive Latent Prediction

Shuyuan Tu, Yueming Pan, Yinming Huang and 6 more

Shuyuan Tu, Yueming Pan, Yinming Huang, Xintong Han, Zhen Xing, Qi Dai, Kai Qiu, Chong Luo, Zuxuan Wu

arXiv 2512.16900 · 202510 upvotes on Hugging Face

Research ideas from this paper (3)

Generated by AI. The paper’s authors did not write these.

AI summary

The paper presents FlashPortrait, a video diffusion transformer framework for generating infinite-length, identity-preserving portrait animations with improved inference speed. It introduces an Adaptive Latent Prediction mechanism that uses Taylor expansion to skip denoising steps and a Normalized Facial Expression Block to align facial features with diffusion latents, achieving up to 6x acceleration compared to baselines.

How it works (AI summary)

The model is based on the Wan2.1 Video Diffusion Transformer (DiT). It employs a Normalized Facial Expression Block to align the distribution statistics (mean/variance) of extracted facial features with diffusion latents. For acceleration, it utilizes a training-free Adaptive Latent Prediction mechanism that applies Taylor series expansion on historical latents within a sliding window to predict future latents, modulated by dynamic functions based on latent variation rates.