AI Paper Atlas

Research map / Image generation / GANs

SESAME: Semantic Editing of Scenes by Adding, Manipulating or Erasing Objects

Evangelos Ntavelis, Andrés Romero, Iason Kastanis, Luc Van Gool, Radu Timofte

ECCV 2020 · Image generation › GANs · approach: Large-scale GAN latent analysis · 75 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 SESAME discriminator in place of PatchGAN (ablation)Hong et al. generator, Full labels · Cityscapes additionmIoU: 60.7 -> 64.0Improved
add SESAME discriminator in place of PatchGAN (ablation)Hong et al. generator, Full labels · Cityscapes additionFID: 12.11 -> 11.76Improved
use SESAME generator in place of Hong et al. generator (ablation)PatchGAN, BBox labels · Cityscapes additionFID: 11.50 -> 11.13Improved
use SESAME discriminator with SESAME generator (ablation)SESAME generator, BBox labels · ADE20k additionFID: 28.16 -> 26.66Improved
use BBox semantics instead of Full semantics (ablation)Hong et al. generator + PatchGAN · Cityscapes additionFID: 12.11 -> 11.50Improved
use BBox semantics instead of Full semantics (ablation)Hong et al. generator + PatchGAN · ADE20k additionFID: 28.48 -> 28.36No clear effect
add SESAME discriminator in place of PatchGAN (ablation)Pix2PixHD generator · Cityscapes label-to-image generationFID: 95.0 -> 55.4Improved
add SESAME discriminator in place of PatchGAN (ablation)SPADE generator · Cityscapes label-to-image generationFID: 71.8 -> 54.2Improved
add SESAME discriminator in place of PatchGAN (ablation)SPADE generator · ADE20k label-to-image generationmIoU: 38.5 -> 49.0Improved
merge streams with concatenation or element-wise product instead of Sum Global Pooling (ablation)SESAME discriminator · Cityscapes additionFID: 11.96 and 12.02Improved
replace SESAME discriminator with PatchGAN for object replacement (ablation)SESAME generator · Cityscapes object replacementFID: 10.63 -> 9.3Improved
add SESAME discriminator with Pix2PixHD++ generator (Full semantics) (ablation)SPADEPix2PixHD generator · Cityscapes additionFID: 11.92 -> 12.74Hurt

Later papers that compared with this method

5 later accepted papers report Sesame as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.

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