AI Paper Atlas

Research map / Detection and segmentation / Weakly supervised detection and segmentation

Image Manipulation Detection With Implicit Neural Representation and Limited Supervision

Zhenfei Zhang, Mingyang Li, Xin Li, Ming-Ching Chang, Jun-Wei Hsieh

ECCV 2024 · Detection and segmentation › Weakly supervised detection and segmentation · approach: Weakly supervised detection · 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 INR reconstruction error map (NRR) as input (ablation)FCN baseline, weakly supervised · CASIAv1P-F1: 0.122 -> 0.159Improved
add selective pixel-level contrastive learning (ablation)Baseline+NRR · CASIAv1P-F1: 0.159 -> 0.166Improved
add adaptive global average pooling (ablation)Baseline+NRR+SCL · CASIAv1P-F1: 0.166 -> 0.199Improved
add NRR, SCL and AGAP to FCN baseline (ablation)FCN · NIST16P-F1: 0.052 -> 0.131Improved
add pseudo-label selection to self-distillation (ablation)unsupervised IMD · NIST16P-F1: 0.082 -> 0.129Improved
use adaptive global-average pooling instead of global-max pooling (ablation)weakly supervised IMD · CASIAv1C-F1: 0.062 -> 0.310Improved
use adaptive pooling instead of global-average pooling (ablation)weakly supervised IMD · CASIAv1C-F1: 0.195 -> 0.310Improved
use weakly supervised method instead of WSCLweakly supervised IMD · ColumbiaI-F1: 0.720 -> 0.695Hurt
train with image-level labels on 12.5K images versus fully supervised methods trained on large synthetic dataweakly supervised IMD · MagicBrush novel manipulationC-F1: 0.382 vs 0.418 (Trufor), 0.324 (PSCC-Net)Mixed
INR-reconstructed test images evaluated by fully supervised methodsfully supervised IMD methods · CASIAv1, Coverage, Columbiapixel-level F1Hurt

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