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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Robust face reconstruction from monocular image in general lightingconditions is challenging. Methods combining deep neural network encoders withdifferentiable rendering have opened up the path for very fast monocularreconstruction of geometry, lighting and reflectance. They can also be trainedin self-supervised manner for increased robustness and better generalization.However, their differentiable rasterization based image formation models, aswell as underlying scene parameterization, limit them to Lambertian facereflectance and to poor shape details. More recently, ray tracing wasintroduced for monocular face reconstruction within a classicoptimization-based framework and enables state-of-the art results. Howeveroptimization-based approaches are inherently slow and lack robustness. In thispaper, we build our work on the aforementioned approaches and propose a newmethod that greatly improves reconstruction quality and robustness in generalscenes. We achieve this by combining a CNN encoder with a differentiable raytracer, which enables us to base the reconstruction on much more advancedpersonalized diffuse and specular albedos, a more sophisticated illuminationmodel and a plausible representation of self-shadows. This enables to take abig leap forward in reconstruction quality of shape, appearance and lightingeven in scenes with difficult illumination. With consistent face attributesreconstruction, our method leads to practical applications such as relightingand self-shadows removal. Compared to state-of-the-art methods, our resultsshow improved accuracy and validity of the approach.

Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray Tracing