English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Constructing vivid 3D head avatars for given subjects and realizing a seriesof animations on them is valuable yet challenging. This paper presentsGaussianHead, which models the actional human head with anisotropic 3DGaussians. In our framework, a motion deformation field and multi-resolutiontri-plane are constructed respectively to deal with the head's dynamic geometryand complex texture. Notably, we impose an exclusive derivation scheme on eachGaussian, which generates its multiple doppelgangers through a set of learnableparameters for position transformation. With this design, we can compactly andaccurately encode the appearance information of Gaussians, even those fittingthe head's particular components with sophisticated structures. In addition, aninherited derivation strategy for newly added Gaussians is adopted tofacilitate training acceleration. Extensive experiments show that our methodcan produce high-fidelity renderings, outperforming state-of-the-art approachesin reconstruction, cross-identity reenactment, and novel view synthesis tasks.Our code is available at: https://github.com/chiehwangs/gaussian-head.

GaussianHead: High-fidelity Head Avatars with Learnable Gaussian Derivation