The Experts below are selected from a list of 7002 Experts worldwide ranked by ideXlab platform

Takeshi Kumaki - One of the best experts on this subject based on the ideXlab platform.

  • Morphological Pattern Spectrum Based Image Manipulation Detection
    2017 IEEE 7th International Advance Computing Conference (IACC), 2020
    Co-Authors: Nayana Nayak, P. Nidhi Hegde, Anusha, Panchami Nayak, P. S. Venugopala, Takeshi Kumaki
    Abstract:

    This paper presents an Image histogram-based Image Manipulation detection method in android. The method consists of using a mathematical morphological-based algorithm to extract pictorial feature information from an original digital Image. This will be useful in situations where it is important to find evidence of specific events such as investigation of crimes or simply Image comparison. The morphological pattern spectrum was implemented in android platform using Java and OpenCV. Analysis of manipulated Images indicated that the proposed detection method was able to clearly identify the differences from the original Images. The results show that the proposed technique has sufficient ability to distinguish the very slight Manipulation of upto one pixel size.

  • ISDCS - Structuring Element-counting Approach for Morphological Pattern Spectrum-based Image Manipulation Detection
    2019 2nd International Symposium on Devices Circuits and Systems (ISDCS), 2019
    Co-Authors: Kyosuke Kageyama, Takeshi Kumaki, Tetsushi Koide
    Abstract:

    The use of digital Images has become quite widespread in legal, medical, and private contexts. However, anyone can easily edit or manipulate any digital Image on a computer. Thus, an effective method for detecting digital Image Manipulation is required for retaining authenticity. Image-Manipulation detection is applied in investigations of crimes and photographic evidence. In this paper, we describe a technique for detecting a manipulated Image by morphological pattern spectrum. Before we have researched morphological pattern spectrum detected Manipulation. We propose the new technique which counts a number of the same scale as the structuring element scale by morphological pattern spectrum in an Image. So it can judge a manipulated Image in detail because the new technique can detect the number of a manipulated scale. Thus, even if the previous technique judge a large different in the pattern spectrum between the original Image and the manipulated Image, the new technique can judge to be actually small Manipulation.

Philip H. S. Torr - One of the best experts on this subject based on the ideXlab platform.

  • ManiGAN: Text-Guided Image Manipulation
    2020 IEEE CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020
    Co-Authors: Bowen Li, Xiaojuan Qi, Thomas Lukasiewicz, Philip H. S. Torr
    Abstract:

    The goal of our paper is to semantically edit parts of an Image matching a given text that describes desired attributes (e.g., texture, colour, and background), while preserving other contents that are irrelevant to the text. To achieve this, we propose a novel generative adversarial network (ManiGAN), which contains two key components: text-Image affine combination module (ACM) and detail correction module (DCM). The ACM selects Image regions relevant to the given text and then correlates the regions with corresponding semantic words for effective Manipulation. Meanwhile, it encodes original Image features to help reconstruct text-irrelevant contents. The DCM rectifies mismatched attributes and completes missing contents of the synthetic Image. Finally, we suggest a new metric for evaluating Image Manipulation results, in terms of both the generation of new attributes and the reconstruction of text-irrelevant contents. Extensive experiments on the CUB and COCO datasets demonstrate the superior performance of the proposed method.

  • ManiGAN: Text-Guided Image Manipulation
    arXiv: Computer Vision and Pattern Recognition, 2019
    Co-Authors: Bowen Li, Xiaojuan Qi, Thomas Lukasiewicz, Philip H. S. Torr
    Abstract:

    The goal of our paper is to semantically edit parts of an Image matching a given text that describes desired attributes (e.g., texture, colour, and background), while preserving other contents that are irrelevant to the text. To achieve this, we propose a novel generative adversarial network (ManiGAN), which contains two key components: text-Image affine combination module (ACM) and detail correction module (DCM). The ACM selects Image regions relevant to the given text and then correlates the regions with corresponding semantic words for effective Manipulation. Meanwhile, it encodes original Image features to help reconstruct text-irrelevant contents. The DCM rectifies mismatched attributes and completes missing contents of the synthetic Image. Finally, we suggest a new metric for evaluating Image Manipulation results, in terms of both the generation of new attributes and the reconstruction of text-irrelevant contents. Extensive experiments on the CUB and COCO datasets demonstrate the superior performance of the proposed method. Code is available at this https URL.

Tetsushi Koide - One of the best experts on this subject based on the ideXlab platform.

  • ISDCS - Structuring Element-counting Approach for Morphological Pattern Spectrum-based Image Manipulation Detection
    2019 2nd International Symposium on Devices Circuits and Systems (ISDCS), 2019
    Co-Authors: Kyosuke Kageyama, Takeshi Kumaki, Tetsushi Koide
    Abstract:

    The use of digital Images has become quite widespread in legal, medical, and private contexts. However, anyone can easily edit or manipulate any digital Image on a computer. Thus, an effective method for detecting digital Image Manipulation is required for retaining authenticity. Image-Manipulation detection is applied in investigations of crimes and photographic evidence. In this paper, we describe a technique for detecting a manipulated Image by morphological pattern spectrum. Before we have researched morphological pattern spectrum detected Manipulation. We propose the new technique which counts a number of the same scale as the structuring element scale by morphological pattern spectrum in an Image. So it can judge a manipulated Image in detail because the new technique can detect the number of a manipulated scale. Thus, even if the previous technique judge a large different in the pattern spectrum between the original Image and the manipulated Image, the new technique can judge to be actually small Manipulation.

Bowen Li - One of the best experts on this subject based on the ideXlab platform.

  • ManiGAN: Text-Guided Image Manipulation
    2020 IEEE CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020
    Co-Authors: Bowen Li, Xiaojuan Qi, Thomas Lukasiewicz, Philip H. S. Torr
    Abstract:

    The goal of our paper is to semantically edit parts of an Image matching a given text that describes desired attributes (e.g., texture, colour, and background), while preserving other contents that are irrelevant to the text. To achieve this, we propose a novel generative adversarial network (ManiGAN), which contains two key components: text-Image affine combination module (ACM) and detail correction module (DCM). The ACM selects Image regions relevant to the given text and then correlates the regions with corresponding semantic words for effective Manipulation. Meanwhile, it encodes original Image features to help reconstruct text-irrelevant contents. The DCM rectifies mismatched attributes and completes missing contents of the synthetic Image. Finally, we suggest a new metric for evaluating Image Manipulation results, in terms of both the generation of new attributes and the reconstruction of text-irrelevant contents. Extensive experiments on the CUB and COCO datasets demonstrate the superior performance of the proposed method.

  • ManiGAN: Text-Guided Image Manipulation
    arXiv: Computer Vision and Pattern Recognition, 2019
    Co-Authors: Bowen Li, Xiaojuan Qi, Thomas Lukasiewicz, Philip H. S. Torr
    Abstract:

    The goal of our paper is to semantically edit parts of an Image matching a given text that describes desired attributes (e.g., texture, colour, and background), while preserving other contents that are irrelevant to the text. To achieve this, we propose a novel generative adversarial network (ManiGAN), which contains two key components: text-Image affine combination module (ACM) and detail correction module (DCM). The ACM selects Image regions relevant to the given text and then correlates the regions with corresponding semantic words for effective Manipulation. Meanwhile, it encodes original Image features to help reconstruct text-irrelevant contents. The DCM rectifies mismatched attributes and completes missing contents of the synthetic Image. Finally, we suggest a new metric for evaluating Image Manipulation results, in terms of both the generation of new attributes and the reconstruction of text-irrelevant contents. Extensive experiments on the CUB and COCO datasets demonstrate the superior performance of the proposed method. Code is available at this https URL.

Mauro Barni - One of the best experts on this subject based on the ideXlab platform.

  • Effectiveness of random deep feature selection for securing Image Manipulation detectors against adversarial examples.
    arXiv: Cryptography and Security, 2019
    Co-Authors: Mauro Barni, Ehsan Nowroozi, Benedetta Tondi, Bowen Zhang
    Abstract:

    We investigate if the random feature selection approach proposed in [1] to improve the robustness of forensic detectors to targeted attacks, can be extended to detectors based on deep learning features. In particular, we study the transferability of adversarial examples targeting an original CNN Image Manipulation detector to other detectors (a fully connected neural network and a linear SVM) that rely on a random subset of the features extracted from the flatten layer of the original network. The results we got by considering three Image Manipulation detection tasks (resizing, median filtering and adaptive histogram equalization), two original network architectures and three classes of attacks, show that feature randomization helps to hinder attack transferability, even if, in some cases, simply changing the architecture of the detector, or even retraining the detector is enough to prevent the transferability of the attacks.

  • secure detection of Image Manipulation by means of random feature selection
    IEEE Transactions on Information Forensics and Security, 2019
    Co-Authors: Zhipeng Chen, Benedetta Tondi, Yao Zhao, Rongrong Ni, Xiaolong Li, Mauro Barni
    Abstract:

    We address the problem of data-driven Image Manipulation detection in the presence of an attacker with limited knowledge about the detector. Specifically, we assume that the attacker knows the architecture of the detector, the training data, and the class of features $\mathcal V$ the detector can rely on. In order to get an advantage in his race of arms with the attacker, the analyst designs the detector by relying on a subset of features chosen at random in $\mathcal V$ . Given its ignorance about the exact feature set, the adversary attacks a version of the detector based on the entire feature set. In this way, the effectiveness of the attack diminishes since there is no guarantee that attacking a detector working in the full feature space will result in a successful attack against the reduced-feature detector. We theoretically prove that, thanks to random feature selection, the security of the detector significantly increases at the expense of a negligible loss of performance in the absence of attacks. We also provide an experimental validation of the proposed procedure by focusing on the detection of two specific kinds of Image Manipulations, namely adaptive histogram equalization and median filtering. The experiments confirm the gain in security at the expense of a negligible loss of performance in the absence of attacks.

  • Secure Detection of Image Manipulation by means of Random Feature Selection
    arXiv: Cryptography and Security, 2018
    Co-Authors: Zhipeng Chen, Benedetta Tondi, Yao Zhao, Rongrong Ni, Xiaolong Li, Mauro Barni
    Abstract:

    We address the problem of data-driven Image Manipulation detection in the presence of an attacker with limited knowledge about the detector. Specifically, we assume that the attacker knows the architecture of the detector, the training data and the class of features V the detector can rely on. In order to get an advantage in his race of arms with the attacker, the analyst designs the detector by relying on a subset of features chosen at random in V. Given its ignorance about the exact feature set, the adversary must attack a version of the detector based on the entire feature set. In this way, the effectiveness of the attack diminishes since there is no guarantee that attacking a detector working in the full feature space will result in a successful attack against the reduced-feature detector. We prove both theoretically and experimentally - by applying the proposed procedure to the detection of two specific kinds of Image Manipulations - that, thanks to random feature selection, the security of the detector increases significantly at the expense of a negligible loss of performance in the absence of attacks. We theoretically prove that, under some simplifying assumptions, the security of the detector increases significantly thanks to random feature selection. We also provide an experimental validation of the proposed procedure by focusing on the detection of two specific kinds of Image Manipulations. The experiments confirm the gain in security at the expense of a negligible loss of performance in the absence of attacks.