AI Paper Atlas

Research map / Image generation / Image restoration

Zero-Shot Learning for Reflection Removal of Single 360-Degree Image

Byeong-Ju Han, Jae-Young Sim

ECCV 2022 · Image generation › Image restoration · approach: Image deblurring

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 gradient-based matching to color-based matching for reference search (ablation)Proposed · aquarium 360-degree images (qualitative)reflection recovery qualityImproved
add color-based matching to gradient-based matching (ablation)Proposed · aquarium 360-degree images (qualitative)reflection color fidelityImproved
add glass synthesis loss Lrecon(I~G, IG) (ablation)Proposed · real 360-degree images (qualitative)transmission sharpnessImproved
add gradient prior loss Lgrad (ablation)Proposed · real 360-degree images (qualitative)reflection edge suppression in transmissionImproved
use Proposed instead of PBTIPBTI · Synthetic 360-degree imagesPSNR-T: 12.70 -> 19.08Improved
use Proposed instead of IBCLNIBCLN · Synthetic 360-degree imagesPSNR-R: 12.28 -> 28.31Improved
use Proposed instead of PRRPRR · Synthetic 360-degree imagesSSIM-T: 0.675 -> 0.647Hurt
increase angular deviation of glass plane orientationProposed · Fig. 9 limitationtransmission reflection removalHurt

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