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

In this paper, we tackle the challenging task of learning a generalizablehuman NeRF model from a monocular video. Although existing generalizable humanNeRFs have achieved impressive results, they require muti-view images or videoswhich might not be always available. On the other hand, some works onfree-viewpoint rendering of human from monocular videos cannot be generalizedto unseen identities. In view of these limitations, we propose GHuNeRF to learna generalizable human NeRF model from a monocular video of the human performer.We first introduce a visibility-aware aggregation scheme to compute vertex-wisefeatures, which is used to construct a 3D feature volume. The feature volumecan only represent the overall geometry of the human performer withinsufficient accuracy due to the limited resolution. To solve this, we furtherenhance the volume feature with temporally aligned point-wise features using anattention mechanism. Finally, the enhanced feature is used for predictingdensity and color for each sampled point. A surface-guided sampling strategy isalso adopted to improve the efficiency for both training and inference. Wevalidate our approach on the widely-used ZJU-MoCap dataset, where we achievecomparable performance with existing multi-view video based approaches. We alsotest on the monocular People-Snapshot dataset and achieve better performancethan existing works when only monocular video is used. Our code is available atthe project website.

GHuNeRF: Generalizable Human NeRF from a Monocular Video