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

Removing soft and self shadows that lack clear boundaries from a single imageis still challenging. Self shadows are shadows that are cast on the objectitself. Most existing methods rely on binary shadow masks, without consideringthe ambiguous boundaries of soft and self shadows. In this paper, we presentDeS3, a method that removes hard, soft and self shadows based on adaptiveattention and ViT similarity. Our novel ViT similarity loss utilizes featuresextracted from a pre-trained Vision Transformer. This loss helps guide thereverse sampling towards recovering scene structures. Our adaptive attention isable to differentiate shadow regions from the underlying objects, as well asshadow regions from the object casting the shadow. This capability enables DeS3to better recover the structures of objects even when they are partiallyoccluded by shadows. Different from existing methods that rely on constraintsduring the training phase, we incorporate the ViT similarity during thesampling stage. Our method outperforms state-of-the-art methods on the SRD,AISTD, LRSS, USR and UIUC datasets, removing hard, soft, and self shadowsrobustly. Specifically, our method outperforms the SOTA method by 16% of theRMSE of the whole image on the LRSS dataset.

DeS3: Adaptive Attention-driven Self and Soft Shadow Removal using ViT Similarity