AI Paper Atlas

Research map / Detection and segmentation / CNN features

Exploiting Spatial Structure for Localizing Manipulated Image Regions

Jawadul H. Bappy, Amit K. Roy-Chowdhury, Jason Bunk, Lakshmanan Nataraj, B. S. Manjunath

ICCV 2017 · Detection and segmentation › CNN features · approach: Forgery detection · 280 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 joint patch classification and segmentation trainingConv-LSTM network · NIST patch classificationpatch accuracy: 86.76% -> 89.38%Improved
add joint trainingConv-LSTM network · IEEE Forensics Challenge patch classificationpatch accuracy: 86.05% -> 87.68%Improved
add joint trainingConv-LSTM network · COVERAGE patch classificationpatch accuracy: 79.12% -> 80.06%Improved
add joint patch classification trainingConv-LSTM-Conv network · NIST segmentationsegmentation accuracy: 82.78% -> 84.60%Improved
add joint patch classification trainingConv-LSTM-Conv network · IEEE Forensics Challenge segmentationsegmentation accuracy: 74.32% -> 77.67%Improved
add joint patch classification trainingConv-LSTM-Conv network · COVERAGE segmentationsegmentation accuracy: 78.42% -> 81.14%Improved
add two convolution layers before the LSTM (ablation)S-LSTM-Conv · NIST segmentationsegmentation accuracy: 78.60% -> 82.78%Improved
add conv5 layer after LSTM (conv3,4 only vs conv3,4,5) (ablation)S-LSTM network · NIST segmentationsegmentation accuracy: 74.43% -> 78.60%Improved
use LSTM-based network instead of FCN (ablation)segmentation network · NIST segmentationsegmentation accuracy: 64.34% -> 74.43%Improved
add a convolutional front end before the LSTM for patch classificationP-LSTM · NIST patch classificationpatch accuracy: 81.05% -> 86.76%Improved
use 16 feature maps in conv1 instead of 8 (ablation)J-Conv-LSTM-Conv · NIST segmentationsegmentation accuracy: 80.88% -> 84.60%Improved
use 16 feature maps in conv1 instead of 32 (ablation)J-Conv-LSTM-Conv · NIST segmentationsegmentation accuracy: 73.17% -> 84.60%Improved

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