Apple releases real-time Gaussian head animation method GHARP
The story so far
Apple Machine Learning Research has released the GHARP method, which animates 3D heads in real-time using a small number of input images and expression signals. The method is divided into two stages: offline construction of geometric and appearance representations, and runtime prediction of expression residuals. Animation is performed in the semantic structure latent space of a pre-trained reconstruction model, achieving efficient residual prediction. GHARP achieves SOTA quality on the Ava-256 benchmark, with a 13x speedup on A100 GPUs and an 8x reduction in Gaussian count. GHARP leverages large-scale reconstruction priors to constrain the space of expression changes, thereby improving animation efficiency. This method, proposed by Apple Machine Learning Research, aims to achieve real-time 3D head animation.
AI-written from the reports · Updated 23 days ago
Coverage timeline
Follow the coverage to see every side of the story.
- Apple Machine Learning ResearchGHARP: Real-Time Gaussian Head Animation from Large-Scale Reconstruction Prior
The GHARP method animates 3D heads in real-time using a small amount of input images and expression signals, divided into an offline identity phase (constructing geometry and appearance representations) and a runtime animation phase (predicting expression residuals). This method performs animation in the semantic structure latent space of the pre-trained reconstruction model, limiting the expression change space to achieve efficient residual prediction, and achieves SOTA quality on the Ava-256 benchmark, with a 13x speedup on A100 GPUs and an 8x reduction in the number of Gaussians.