Papers › Constrained R-CNN: A general image manipulation detection model

Constrained R-CNN: A general image manipulation detection model

19 Nov 2019arXiv:1911.08217archive 2025-07-28

Chao Yang, Huizhou Li, Fangting Lin, Bin Jiang, Hao Zhao

Recently, deep learning-based models have exhibited remarkable performance for image manipulation detection. However, most of them suffer from poor universality of handcrafted or predetermined features. Meanwhile, they only focus on manipulation localization and overlook manipulation classification. To address these issues, we propose a coarse-to-fine architecture named Constrained R-CNN for complete and accurate image forensics. First, the learnable manipulation feature extractor learns a unified feature representation directly from data. Second, the attention region proposal network effectively discriminates manipulated regions for the next manipulation classification and coarse localization. Then, the skip structure fuses low-level and high-level information to refine the global manipulation features. Finally, the coarse localization information guides the model to further learn the finer local features and segment out the tampered region. Experimental results show that our model achieves state-of-the-art performance. Especially, the F1 score is increased by 28.4%, 73.2%, 13.3% on the NIST16, COVERAGE, and Columbia dataset.

PaperPDF

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

General ClassificationImage ForensicsImage ManipulationImage Manipulation DetectionImage Manipulation LocalizationRegion Proposal

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Image Manipulation Detection COVERAGE CR-CNN AUC .553 #7 of 8 Archive leaderboard report
Image Manipulation Detection COVERAGE CR-CNN Balanced Accuracy .391 #7 of 8 Archive leaderboard report
Image Manipulation Detection Casia V1+ CR-CNN AUC .670 #7 of 9 Archive leaderboard report
Image Manipulation Detection Casia V1+ CR-CNN Balanced Accuracy .481 #7 of 9 Archive leaderboard report
Image Manipulation Detection CocoGlide CR-CNN AUC .589 #6 of 8 Archive leaderboard report
Image Manipulation Detection CocoGlide CR-CNN Balanced Accuracy .447 #6 of 8 Archive leaderboard report
Image Manipulation Detection Columbia CR-CNN AUC .755 #6 of 8 Archive leaderboard report
Image Manipulation Detection Columbia CR-CNN Balanced Accuracy .631 #6 of 8 Archive leaderboard report
Image Manipulation Detection DSO-1 CR-CNN AUC .576 #7 of 9 Archive leaderboard report
Image Manipulation Detection DSO-1 CR-CNN Balanced Accuracy .289 #7 of 9 Archive leaderboard report
Image Manipulation Localization COVERAGE CR-CNN Average Pixel F1(Fixed threshold) .391 #8 of 11 Archive leaderboard report
Image Manipulation Localization Casia V1+ CR-CNN Average Pixel F1(Fixed threshold) .481 #9 of 11 Archive leaderboard report
Image Manipulation Localization CocoGlide CR-CNN Average Pixel F1(Fixed threshold) .447 #9 of 11 Archive leaderboard report
Image Manipulation Localization Columbia CR-CNN Average Pixel F1(Fixed threshold) .631 #10 of 11 Archive leaderboard report
Image Manipulation Localization DSO-1 CR-CNN Average Pixel F1(Fixed threshold) .289 #10 of 11 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections