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 aim at synthesizing a free-viewpoint video of an arbitraryhuman performance using sparse multi-view cameras. Recently, several works haveaddressed this problem by learning person-specific neural radiance fields(NeRF) to capture the appearance of a particular human. In parallel, some workproposed to use pixel-aligned features to generalize radiance fields toarbitrary new scenes and objects. Adopting such generalization approaches tohumans, however, is highly challenging due to the heavy occlusions and dynamicarticulations of body parts. To tackle this, we propose Neural Human Performer,a novel approach that learns generalizable neural radiance fields based on aparametric human body model for robust performance capture. Specifically, wefirst introduce a temporal transformer that aggregates tracked visual featuresbased on the skeletal body motion over time. Moreover, a multi-view transformeris proposed to perform cross-attention between the temporally-fused featuresand the pixel-aligned features at each time step to integrate observations onthe fly from multiple views. Experiments on the ZJU-MoCap and AIST datasetsshow that our method significantly outperforms recent generalizable NeRFmethods on unseen identities and poses. The video results and code areavailable at https://youngjoongunc.github.io/nhp.

Neural Human Performer: Learning Generalizable Radiance Fields for Human Performance Rendering