Image Manipulation Detection by Multi-View Multi-Scale Supervision

Chen, Xinru; Chengbo, Dong,; Ji, Jiaqi; Cao, Juan; Li, Xirong

doi:10.1109/iccv48922.2021.01392

Cited by 147 publications

(49 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, most forgery detection methods focus on identifying a specific tampering type, such as copy‐move 18,22 and splicing 23,24 . To solve these problems, some general works 25–29 begin to study the suppression of image content. Image content suppression means that the network can learn the discrepancy between tampered and non‐tampered regions based on obtaining appropriate semantic information, and then precisely locate the manipulated regions to prevent false localization of candidate regions.…”

Section: Introductionmentioning

confidence: 99%

“…Therefore, GSR‐Net 35 proposes a boundary branch to reconstruct the edges of the region. MVSS‐Net 28 adopts a combination of the boundary branch and a noise view to localize the tamper regions. However, these methods pre‐localize the manipulated regions with the entire image or a bounding box, that inevitably comprises authentic regions hence causes false alarms.…”

Section: Introductionmentioning

confidence: 99%

“…CR‐CNN 25 utilizes constrained convolution in the initial layers to reach the purpose of suppressing image content and Mask R‐CNN 30 to locate rough results in subsequent forgery detection. This noise‐based view 25,27,28,31,32 assumes that the pristine and tampered regions in the splicing image have different noise distributions, which are not valid for copy‐move originating from the same image. Besides, the use of boundary artifacts to identify tampered regions is evolving.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

PR‐NET: Progressively‐refined neural network for image manipulation localization

Shi

Chang

Chen

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

Current deep learning‐based image manipulation localization methods achieve impressive performance when rich spatial features and information are fully utilized. However, most of them suffer from the irrelevance of semantic awareness when identifying various manipulation categories. This leads to false alarms on recognizing forged regions. In this paper, we propose a Progressively‐Refined Neural Network (PR‐Net), to localize the tampered regions progressively under a coarse‐to‐fine workflow. Specifically, PR‐Net is composed of a Feature Extractor (FE) that captures feature intrinsic correlations and a Mask Generation Module (MGM) with three refining generators. The FE takes a CNN to extract the image features and introduces an attention mechanism Convolution Block Attention Module (CBAM) to suppress the image content and guide the extractor in exploring the inconsistencies between the manipulated and authentic regions. The MGM comprises three generators where the Coarse Mask RR‐Generator generates a localization result roughly, the Candidate Mask RR‐Generator generates a possible tampered region according to the rough localization measure, and the Fine Mask RR‐Generator produces the final prediction of manipulated regions. We also utilize the Rotated Residual (RR) structure to suppress the image content during the generative process. The extensive experimental results on four benchmark data sets (NIST16, COVER, CASIA v1.0, and In‐The‐Wild) demonstrate the superior performance of PR‐Net compared with the state‐of‐the‐art methods in localizing the manipulated regions.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

PR‐NET: Progressively‐refined neural network for image manipulation localization

Shi

Chang

Chen

et al. 2022

Int J of Intelligent Sys

View full text Add to dashboard Cite

show abstract

“…A preliminary version of this work was published at ICCV 2021 [27]. The journal article improves over the conference paper in multiple aspects.…”

Section: Introductionmentioning

confidence: 99%

“…The journal article improves over the conference paper in multiple aspects. First, for converting pixel-level manipulation detection to an image-level prediction, we propose ConvGeM to replace global max pooling (GMP) used in [27]. The new module effectively overcomes two downsides of GMP, i.e.…”

Section: Introductionmentioning

confidence: 99%

MVSS-Net: Multi-View Multi-Scale Supervised Networks for Image Manipulation Detection

Dong,

Chen,

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The key research question for image manipulation detection is how to learn generalizable features that are sensitive to manipulations in novel data, whilst specific to prevent false alarms on authentic images. Current research emphasizes the sensitivity, with the specificity mostly ignored. In this paper we address both aspects by multi-view feature learning and multi-scale supervision. By exploiting noise distribution and boundary artifacts surrounding tampered regions, the former aims to learn semantic-agnostic and thus more generalizable features. The latter allows us to learn from authentic images which are nontrivial to be taken into account by the prior art that relies on a semantic segmentation loss. Our thoughts are realized by a new network which we term MVSS-Net and its enhanced version MVSS-Net++. Comprehensive experiments on six public benchmark datasets justify the viability of the MVSS-Net series for both pixel-level and image-level manipulation detection.

show abstract

Rethinking Boundary Detection in Deep Learning Models for Medical Image Segmentation

Lin

Zhang

Xiao

et al. 2023

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Image Manipulation Detection by Multi-View Multi-Scale Supervision

Cited by 147 publications

References 20 publications

PR‐NET: Progressively‐refined neural network for image manipulation localization

PR‐NET: Progressively‐refined neural network for image manipulation localization

MVSS-Net: Multi-View Multi-Scale Supervised Networks for Image Manipulation Detection

Rethinking Boundary Detection in Deep Learning Models for Medical Image Segmentation

Contact Info

Product

Resources

About