AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Multi-Target Adversarial Frameworks for Domain Adaptation in Semantic Segmentation

Antoine Saporta, Tuan-Hung Vu, Matthieu Cord, Patrick Pérez

ICCV 2021 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Adaptation for semantic segmentation · 50 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
use MTKT multi-teacher/single-student distillation instead of merging targetsMulti-Target Baseline · GTA5->Cityscapes+MapillarymIoU Avg.: +2.0%Improved
use multi-discriminator alignment instead of merging targetsMulti-Target Baseline · GTA5->Cityscapes+MapillarymIoU Avg.: +0.6%Improved
use multi-discriminator instead of merged-target trainingMulti-Target Baseline · GTA5->Cityscapes+IDDmIoU Avg.: 67.4 -> 67.3No clear effect
use MTKT instead of merged-target trainingMulti-Target Baseline · GTA5->Cityscapes+Mapillary+IDDmIoU Avg.: 67.8 -> 69.1Improved
use MTKT instead of merged-target trainingMulti-Target Baseline · Cityscapes->Mapillary+IDDmIoU Avg.: 67.0 -> 69.0Improved
train on multiple target domains instead of a single targetAdvEnt single-target · GTA5->Cityscapes+MapillarymIoU Avg.: 66.6 -> 68.9Improved
add adversarial training on the target-agnostic branch of MTKT (ablation)MTKT · GTA5->Cityscapes+MapillarymIoU Avg.: -0.9%Hurt
add pseudo-labeling (ESL) to MTKT, refining target-specific head only (ablation)MTKT · GTA5->Cityscapes+IDDmIoU Avg.: 68.2 -> 69.8Improved
add pseudo-labeling (ESL) to multi-target baseline (ablation)Multi-Target Baseline · GTA5->Cityscapes+IDDmIoU Avg.: 67.4 -> 68.9Improved
refine target-agnostic model too with pseudo-labels (PL 3) rather than head only (PL 1) (ablation)MTKT + PL · GTA5->Cityscapes+IDDmIoU Avg.: 69.8 -> 69.9No clear effect
use MTKT instead of multi-target baseline for direct transfer to unseen domainMTKT · Direct transfer to MapillarymIoU: 66.7 -> 70.6Improved
use multi-discriminator instead of multi-target baseline for direct transfer to unseen domainMulti-Dis. · Direct transfer to MapillarymIoU: 66.7 -> 65.9Hurt

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