AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Perceptual Artifacts Localization for Inpainting

Lingzhi Zhang, Yuqian Zhou, Connelly Barnes, Sohrab Amirghodsi, Zhe Lin, Eli Shechtman, Jianbo Shi

ECCV 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning · 20 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 the perfectly filled images (no artifact labels) to training (ablation)PSPNet segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 43.83 -> 46.04Improved
start from ADE20K pretrained weights (ablation)PSPNet segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 44.93 -> 46.04Improved
concatenate the hole mask to the input (ablation)PSPNet segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 46.04 -> 45.96No clear effect
add pseudo-label pretraining on 100K unlabeled images (ablation)PSPNet segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 46.04 -> 46.44Improved
add real images with empty masks to pseudo pretraining (ablation)PSPNet segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 46.44 -> 46.77Improved
use Swin-B backbone with UperNet head instead of ResNet-50 and PSPNet (ablation)segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 46.04 -> 44.20Hurt
use HRNet head instead of PSPNet head (ablation)segmenter · Perceptual artifacts segmentation (test, 479 images)IoU: 46.04 -> 41.35Hurt
use PAR instead of PSNR to rank inpainting methods vs human preferencePAR metric · Places2 object removal pairs (1966 strongly preferred pairs)agreement with human preference: 41.50% -> 72.89%Improved
use PAR instead of MUSIQ to rank inpainting methods vs human preferencePAR metric · Places2 object removal pairs (1966 strongly preferred pairs)agreement with human preference: 61.28% -> 72.89%Improved
use PAR instead of LPIPSPAR metric · Places2 object removal pairs (1966 strongly preferred pairs)agreement with human preference: 38.55% -> 72.89%Improved
add iterative fill on predicted artifact regions (5th refill) to EdgeConnectEdgeConnect · User study (500 images)user preference: 10.6% preferred original -> 37.8% preferred iterativeImproved
add iterative fill on predicted artifact regions (5th refill) to LaMaLaMa · User study (500 images)user preference: 1.8% preferred original -> 30.0% preferred iterativeImproved

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