AI Paper Atlas

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

SARIF: Segment Anything for Robust Image Forensics

Dong-Hyun Moon, Ju-Hyeon Nam, Sang-Chul Lee

ECCV 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Domain-generalized segmentation · 0 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 LoRA adapters to the SAM image encoder (ablation)vanilla SAM · CASIAv2-trained, unseen averageDSC: 23.3 -> 33.0Improved
add FSIE (dual-encoder residual cues) on top of adapters (ablation)SAM + Adapter · CASIAv2-trained, unseen averageDSC: 33.0 -> 44.7Improved
add FGMD (feedback mask decoder) on top of adapters (ablation)SAM + Adapter · CASIAv2-trained, unseen averageDSC: 33.0 -> 42.6Improved
add FGMD to adapter plus FSIE (ablation)SAM + Adapter + FSIE · CASIAv2-trained, unseen averageDSC: 44.7 -> 57.9Improved
add FSIE to adapter plus FGMD (ablation)SAM + Adapter + FGMD · CASIAv2-trained, unseen averageDSC: 42.6 -> 57.9Improved
add FGMD to adapter plus FSIE (ablation)SAM + Adapter + FSIE · CASIAv2 seen domainDSC: 62.4 -> 63.1No clear effect
add FSIE to adapter (ablation)SAM + Adapter · FLOPsFLOPs(G): 503.79 -> 994.88Hurt
progressive feedback refinement from initial mask to final mask (ablation)SARIF · CASIAv2 seenIoU/DSC: 53.3/61.9 -> 56.7/63.1Improved
progressive feedback refinement from initial mask to final mask (ablation)SARIF · IMD2020IoU/DSC: 38.2/46.6 -> 40.4/48.4Improved
use SAM-based SARIF instead of M2SFormerSARIF · CocoGlideDSC: 12.2 -> 34.0Improved
use SARIF instead of SAFIRESARIF · Unseen datasets DSCDSC: 65.4 vs 39.5 on CoMoFoD; 48.4 vs 53.1 on IMD2020Mixed
use SARIF instead of FBINet on authentic imagesSARIF · CASIAv2 authenticFPR: 3.0 -> 5.1Hurt

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