AI Paper Atlas

Research map / Image generation / Image restoration

Deep Edge-Aware Interactive Colorization Against Color-Bleeding Effects

Eungyeup Kim, Sanghyeon Lee, Jeonghoon Park, Somi Choi, Choonghyun Seo, Jaegul Choo

ICCV 2021 · Image generation › Image restoration · approach: Image harmonization and color transformation · 38 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 edge-enhancing network with user scribblesZhang et al. (conditional) · ImageNet ctest10k local kernel 7PSNR: 14.966 -> 16.041Improved
add edge-enhancing network with user scribblesZhang et al. (conditional) · ImageNet ctest10k local kernel 7LPIPS: 0.208 -> 0.177Improved
add edge-enhancing network with user scribblesSu et al. (conditional) · ImageNet ctest10k local kernel 7PSNR: 16.393 -> 16.507Improved
add edge-enhancing network with user scribblesZhang et al. (unconditional) · ImageNet ctest10k full imagePSNR: 21.981 -> 22.026No clear effect
add edge-enhancing network with user scribblesZhang et al. (conditional) · ImageNet ctest10kcluster discrepancy ratio: 0.418 -> 0.543Improved
add edge-enhancing network with user scribblesSu et al. (conditional) · ImageNet ctest10kcluster discrepancy ratio: 0.336 -> 0.394Improved
add consistency loss L_con (ablation)edge-enhancing network · ImageNet ctest10k ablationPSNR: 15.799 -> 16.041Improved
add consistency loss L_con (ablation)edge-enhancing network · ImageNet ctest10k ablationcluster discrepancy ratio: 0.472 -> 0.512Mixed
add regularization loss L_reg (ablation)edge-enhancing network · ImageNet ctest10k ablationPSNR: 15.624 -> 16.041Improved
add regularization loss L_reg (ablation)edge-enhancing network · ImageNet ctest10k ablationcluster discrepancy ratio: 0.605 -> 0.512Hurt
add scribble width augmentation (ablation)edge-enhancing network · ImageNet ctest10k ablationPSNR: 13.372 -> 16.041Improved
add scribble width augmentation (ablation)edge-enhancing network · ImageNet ctest10k ablationLPIPS: 0.259 -> 0.177Improved

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