AI Paper Atlas

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

Structure-Guided Image Completion With Image-Level and Object-Level Semantic Discriminators

Haitian Zheng, Zhe Lin, Jingwan Lu, Scott D. Cohen, Eli Shechtman, Connelly Barnes, Jianming Zhang, Qing Liu, Sohrab Amirghodsi, Yuqian Zhou, Jiebo Luo

TPAMI 2024 · Detection and segmentation › Open-set and few-shot detection · approach: Object-level representation learning

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 perceptual loss to adversarial-only training (ablation)conditional CM-GAN baseline · Places2-person (CoModGAN masks)FID: 10.5587 -> 2.8470Improved
add perceptual loss to adversarial-only training (ablation)conditional CM-GAN baseline · Places2-person (object masks)FID: 9.3800 -> 2.7246Improved
add image-level semantic discriminator (ablation)adv. + perc. model · Places2-person (CoModGAN masks)FID: 2.8470 -> 2.2705Improved
add image-level semantic discriminator (ablation)adv. + perc. model · Places2-person (object masks)FID: 2.7246 -> 2.1636Improved
add object-level discriminator alone (ablation)adv. + perc. model · Places2-person (CoModGAN masks)FID: 2.8470 -> 2.8100No clear effect
add object-level discriminator alone (ablation)adv. + perc. model · Places2-person (object masks)FID: 2.7246 -> 2.3477Improved
add object-level discriminator to semantic discriminator (ablation)adv. + perc. + sem. D · Places2-person (CoModGAN masks)FID: 2.2705 -> 2.0720Improved
add object-level discriminator to semantic discriminator (ablation)adv. + perc. + sem. D · Places2-person (CoModGAN masks)P-IDS (%): 22.30 -> 25.90Improved
use instance segmentation guidance instead of semantic segmentation (ablation)full model · Places2-person (CoModGAN masks)FID: 2.3860 -> 2.0720Improved
use Canny edge instead of instance edge guidancefull model · Places2-person (CoModGAN masks)FID: 2.6909 -> 1.6736Improved
use our model instead of ControlNet+RePaintControlNet*+RePaint · COCO-Stuff segmentation-guided (object masks)FID: 9.9252 -> 10.1776Hurt
use our one-step GAN instead of diffusion ControlNet+RePaintControlNet+RePaint · inference timeinference time: 3 min 25 sec -> 0.279 secImproved

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