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/

We present TexMesh, a novel approach to reconstruct detailed human mesheswith high-resolution full-body texture from RGB-D video. TexMesh enables highquality free-viewpoint rendering of humans. Given the RGB frames, the capturedenvironment map, and the coarse per-frame human mesh from RGB-D tracking, ourmethod reconstructs spatiotemporally consistent and detailed per-frame meshesalong with a high-resolution albedo texture. By using the incident illuminationwe are able to accurately estimate local surface geometry and albedo, whichallows us to further use photometric constraints to adapt a syntheticallytrained model to real-world sequences in a self-supervised manner for detailedsurface geometry and high-resolution texture estimation. In practice, we trainour models on a short example sequence for self-adaptation and the model runsat interactive framerate afterwards. We validate TexMesh on synthetic andreal-world data, and show it outperforms the state of art quantitatively andqualitatively.

TexMesh: Reconstructing Detailed Human Texture and Geometry from RGB-D Video