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

One crucial aspect of 3D head avatar reconstruction lies in the details offacial expressions. Although recent NeRF-based photo-realistic 3D head avatarmethods achieve high-quality avatar rendering, they still encounter challengesretaining intricate facial expression details because they overlook thepotential of specific expression variations at different spatial positions whenconditioning the radiance field. Motivated by this observation, we introduce anovel Spatially-Varying Expression (SVE) conditioning. The SVE can be obtainedby a simple MLP-based generation network, encompassing both spatial positionalfeatures and global expression information. Benefiting from rich and diverseinformation of the SVE at different positions, the proposed SVE-conditionedneural radiance field can deal with intricate facial expressions and achieverealistic rendering and geometry details of high-fidelity 3D head avatars.Additionally, to further elevate the geometric and rendering quality, weintroduce a new coarse-to-fine training strategy, including a geometryinitialization strategy at the coarse stage and an adaptive importance samplingstrategy at the fine stage. Extensive experiments indicate that our methodoutperforms other state-of-the-art (SOTA) methods in rendering and geometryquality on mobile phone-collected and public datasets.

High-Fidelity 3D Head Avatars Reconstruction through Spatially-Varying Expression Conditioned Neural Radiance Field