AI Paper Atlas

Research map / Image generation / Editing and style transfer

DC-ShadowNet: Single-Image Hard and Soft Shadow Removal Using Unsupervised Domain-Classifier Guided Network

Yeying Jin, Aashish Sharma, Robby T. Tan

ICCV 2021 · Image generation › Editing and style transfer · approach: Image-to-image translation · 171 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add boundary smoothness loss (ablation)DC-ShadowNet · SRDRMSE all: 4.72 -> 4.66Improved
add shadow-robust feature loss (ablation)DC-ShadowNet · SRDRMSE all: 4.83 -> 4.66Improved
add shadow-free chromaticity loss (ablation)DC-ShadowNet · SRDRMSE all: 5.05 -> 4.66Improved
add domain classifier to the generator (ablation)DC-ShadowNet · SRDRMSE all: 5.20 -> 4.66Improved
add domain classifier to the discriminator (ablation)DC-ShadowNet · SRDRMSE all: 5.49 -> 4.66Improved
add both domain classifiers (ablation)DC-ShadowNet · SRDRMSE shadow: 16.10 -> 7.70Improved
add both domain classifiers (ablation)DC-ShadowNet · SRDRMSE non-shadow: 4.80 -> 3.39Improved
add boundary smoothness loss (ablation)DC-ShadowNet · SRDRMSE shadow: 7.80 -> 7.70Improved
add test-time training on test images with unsupervised lossesDC-ShadowNet · LRSSRMSE: 3.48 -> 3.36Improved
add test-time training on test images with unsupervised lossesDC-ShadowNet · LRSSPSNR: 31.01 -> 31.31Improved
use DC-ShadowNet instead of Mask-ShadowGAN unpaired baselineunpaired shadow removal · SRDRMSE shadow: 11.46 -> 7.70Improved
use DC-ShadowNet instead of Mask-ShadowGAN unpaired baselineunpaired shadow removal · AISTDRMSE shadow: 12.5 -> 10.3Improved

Later papers that compared with this method

1 later accepted paper reports DCShadowNet as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy