Image Manipulation Detection On Coverage

Metrics

AUC

Balanced Accuracy

Results

Performance results of various models on this benchmark

			Paper Title
Early Fusion	.839	.770	MMFusion: Combining Image Forensic Filters for Visual Manipulation Detection and Localization
Late Fusion	.792	.720	MMFusion: Combining Image Forensic Filters for Visual Manipulation Detection and Localization
TruFor	.770	.680	TruFor: Leveraging all-round clues for trustworthy image forgery detection and localization
ManTraNet	.760	.500	ManTra-Net: Manipulation Tracing Network for Detection and Localization of Image Forgeries With Anomalous Features
MVSS-Net	.733	.514	Image Manipulation Detection by Multi-View Multi-Scale Supervision
CAT-Net v2	.680	.635	Learning JPEG Compression Artifacts for Image Manipulation Detection and Localization
SPAN	.670	.235	SPAN: Spatial Pyramid Attention Network for Image Manipulation Localization
CR-CNN	.553	.391	Constrained R-CNN: A general image manipulation detection model

0 of 8 row(s) selected.

Image Manipulation Detection On Coverage | SOTA | HyperAI