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

Vision in adverse weather conditions, whether it be snow, rain, or fog ischallenging. In these scenarios, scattering and attenuation severly degradesimage quality. Handling such inclement weather conditions, however, isessential to operate autonomous vehicles, drones and robotic applications wherehuman performance is impeded the most. A large body of work explores removingweather-induced image degradations with dehazing methods. Most methods rely onsingle images as input and struggle to generalize from syntheticfully-supervised training approaches or to generate high fidelity results fromunpaired real-world datasets. With data as bottleneck and most of today'straining data relying on good weather conditions with inclement weather asoutlier, we rely on an inverse rendering approach to reconstruct the scenecontent. We introduce ScatterNeRF, a neural rendering method which adequatelyrenders foggy scenes and decomposes the fog-free background from theparticipating media-exploiting the multiple views from a short automotivesequence without the need for a large training data corpus. Instead, therendering approach is optimized on the multi-view scene itself, which can betypically captured by an autonomous vehicle, robot or drone during operation.Specifically, we propose a disentangled representation for the scatteringvolume and the scene objects, and learn the scene reconstruction withphysics-inspired losses. We validate our method by capturing multi-viewIn-the-Wild data and controlled captures in a large-scale fog chamber.

ScatterNeRF: Seeing Through Fog with Physically-Based Inverse Neural Rendering