AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

SynTag: Enhancing the Geometric Robustness of Inversion-based Generative Image Watermarking

Han Fang, Kejiang Chen, Zehua Ma, Jiajun Deng, Yicong Li, Weiming Zhang, Ee-Chien Chang

ICCV 2025 · Image generation › Autoregressive and latent generation · approach: Robust diffusion watermarking · 1 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 SynTag injection and correctionGauShad · Geometric distortions SD-v1.4TPR: 0.016 -> 0.980Improved
add SynTag injection and correctionTreeRings · Geometric distortions SD-v1.4TPR: 0.548 -> 0.928Improved
add SynTag injectionGauShad · Geometric distortions SD-v1.4bit accuracy: 0.633 -> 0.938Improved
add SynTag injectionGauShad · Non-geometric distortions SD-v1.4TPR: 0.963 -> 0.967No clear effect
add SynTag injectionGauShad · Image qualityFID: 25.23 -> 25.21No clear effect
add SynTag decoder with predictor instead of passive predictor only (ablation)geometric correction · 4 geometric distortionsbit accuracy: 0.730 -> 0.938Improved
add SynTag predictor PSyn (ablation)GauShad-SynTag · 4 geometric distortionsTPR: 0.016 -> 0.763Improved
add SynTag predictor PSyn (ablation)GauShad-SynTag · 4 geometric distortionsbit accuracy: 0.633 -> 0.819Improved
add pixel-level dither compensation Cp (ablation)PSyn · 4 geometric distortionsTPR: 0.763 -> 0.795Improved
add latent-level dither compensation Cl (ablation)PSyn · 4 geometric distortionsTPR: 0.763 -> 0.930Improved
add both Cp and Cl (ablation)PSyn · 4 geometric distortionsTPR: 0.763 -> 0.990Improved
increase purification attack strength from 0.3 to 0.7 (ablation)GauShad-SynTag · Purification attackTPR: 0.980 -> 0.000Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

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