The versatility of diffusion models in generating customized images has led to unauthorized usage of personal artwork, which poses a significant threat to the intellectual property of artists. Existing approaches relying on embedding additional information, such as perturbations, watermarks, and backdoors, suffer from limited defensive capabilities and fail to protect artwork published online. In this paper, we propose StyleSentinel, an approach for copyright protection of artwork by verifying an inherent stylistic fingerprint in the artist's artwork. Specifically, we employ a semantic self-reconstruction process to enhance stylistic expressiveness within the artwork, which establishes a dense and style-consistent manifold foundation for feature learning. Subsequently, we adaptively fuse multi-layer image features to encode abstract artistic style into a compact stylistic fingerprint. Finally, we model the target artist's style as a minimal enclosing hypersphere boundary in the feature space, transforming complex copyright verification into a robust one-class learning task. Extensive experiments demonstrate that compared with the state-of-the-art, StyleSentinel achieves superior performance on the one-sample verification task. We also demonstrate the effectiveness through online platforms.

Diffusion Models (DMs) have achieved remarkable success in realistic voice cloning (VC), while they also increase the risk of malicious misuse. Existing proactive defenses designed for traditional VC models aim to disrupt the forgery process, but they have been proven incompatible with DMs due to the intricate generative mechanisms of diffusion. To bridge this gap, we introduce VoiceCloak, a multi-dimensional proactive defense framework with the goal of obfuscating speaker identity and degrading perceptual quality in potential unauthorized VC. To achieve these goals, we conduct a focused analysis to identify specific vulnerabilities within DMs, allowing VoiceCloak to disrupt the cloning process by introducing adversarial perturbations into the reference audio. Specifically, to obfuscate speaker identity, VoiceCloak first targets speaker identity by distorting representation learning embeddings to maximize identity variation, which is guided by auditory perception principles. Additionally, VoiceCloak disrupts crucial conditional guidance processes, particularly attention context, thereby preventing the alignment of vocal characteristics that are essential for achieving convincing cloning. Then, to address the second objective, VoiceCloak introduces score magnitude amplification to actively steer the reverse trajectory away from the generation of high-quality speech. Noise-guided semantic corruption is further employed to disrupt structural speech semantics captured by DMs, degrading output quality. Extensive experiments highlight VoiceCloak's outstanding defense success rate against unauthorized diffusion-based voice cloning. Additional audio samples of VoiceCloak are available in demo pages.

Image forgery localization aims to precisely identify tampered regions within images, but it commonly depends on costly pixel-level annotations. To alleviate this annotation burden, weakly supervised image forgery localization (WSIFL) has emerged, yet existing methods still achieve limited localization performance as they mainly exploit intra-image consistency clues and lack external semantic guidance to compensate for insufficient supervision information. In this paper, we propose ViLaCo, a vision-language collaborative reasoning framework that introduces auxiliary semantic supervision derived from pre-trained vision-language models (VLMs), enabling accurate pixel-level localization using only image-level labels. Specifically, we first employ a vision-language feature modeling network to jointly extract textual semantics and visual features by leveraging pre-trained VLMs. Next, an adaptive vision-language reasoning network aligns these features through mutual interactions, producing semantically aligned representations. Subsequently, these representations are passed into dual prediction heads, where the coarse head performs image-level classification and the fine head generates pixel-level localization masks, allowing the coarse-grained task to provide guidance for the fine-grained localization. Moreover, a contrastive patch consistency module is introduced to cluster tampered features while separating authentic ones, facilitating more reliable forgery discrimination. Extensive experiments on multiple public datasets demonstrate that ViLaCo substantially outperforms existing WSIFL methods, achieving state-of-the-art performance in both detection and localization accuracy.

With the rapid rise of large models, copyright protection for generated image content has become a critical security challenge. Although deep learning watermarking techniques offer an effective solution for digital image copyright protection, they still face limitations in terms of visual quality, robustness and generalization. To address these issues, this paper proposes an adaptive robust iterative watermarking framework (ARIW-Framework) that achieves high-quality watermarked images while maintaining exceptional robustness and generalization performance. Specifically, we introduce an iterative approach to optimize the encoder for generating robust residuals. The encoder incorporates noise layers and a decoder to compute robustness weights for residuals under various noise attacks. By employing a parallel optimization strategy, the framework enhances robustness against multiple types of noise attacks. Furthermore, we leverage image gradients to determine the embedding strength at each pixel location, significantly improving the visual quality of the watermarked images. Extensive experiments demonstrate that the proposed method achieves superior visual quality while exhibiting remarkable robustness and generalization against noise attacks.

Abstract Artificial intelligence technology based on deep learning has been widely used in key fields such as automatic driving, medical diagnosis and financial risk control. These applications also bring more and more serious security problems. In particular, as the means of attack continue to evolve, well-designed countermeasures seriously threaten the reliability of the model and the security of the system. In order to deal with this risk, defensive confrontation samples have become the core task of AI security research, playing a key role in improving the security and credibility of the model. Aiming at the problems of unclear concepts and overlapping standards in previous classification methods, this paper proposes a clearer and unified classification framework, combs and defines the contents of existing research, and solves the inconsistencies. The framework systematically divides the existing countermeasures and defense methods into three categories: detection, purification and optimization. This classification will help researchers understand the actual effects of different methods in the face of various attacks more clearly. This paper also analyzes the tradeoffs between accuracy, robustness, operational efficiency and generalization capability of various defense mechanisms, and reveals how they balance the calculation cost and actual deployment requirements. In addition, the paper points out the main challenges facing the current research, and puts forward the future research directions, including developing more efficient, adaptive, and cross modal defense methods to comprehensively improve the security of AI systems. The purpose of this review is to help researchers understand the development process of anti sample defense technology and provide a reference path for building a stable and reliable AI system.

The misuse of deep learning-based facial manipulation poses a serious threat to civil rights. To prevent such fraud at its source, proactive defense methods have been proposed that embed invisible adversarial perturbations into images, disrupting the manipulation process and rendering the forged output unconvincing to observers. However, non-targeted disruption of the output may leave identifiable facial features intact, potentially leading to the stigmatization of individuals. In this work, we propose a universal framework for combating facial manipulation, termed ID-Guard. The framework employs a single forward pass of an encoder-decoder network to generate cross-model transferable adversarial perturbations. We introduce a novel Identity Destruction Module (IDM) to suppress identifiable features in manipulated faces. The perturbation generation is optimized by formulating the disruption of various manipulation types as a multi-task learning problem, with a dynamic weighting strategy designed to enhance cross-model performance. Experimental results show that ID-Guard effectively defends against diverse facial manipulation models while degrading identifiable regions in manipulated images. It also enables disrupted images to evade facial inpainting and facial recognition systems. Moreover, ID-Guard can be seamlessly integrated as a plug-and-play component into other tasks, such as adversarial training.

Deepfakes pose significant societal risks, motivating the development of proactive defenses that embed adversarial perturbations in facial images to prevent manipulation. However, in this paper, we show that these preemptive defenses often lack robustness and reliability. We propose a novel approach, Low-Rank Adaptation (LoRA) patching, which injects a plug and-play LoRA patch into Deepfake generators to bypass state of-the-art defenses. A learnable gating mechanism adaptively controls the effect of the LoRA patch and prevents gradient explosions during fine-tuning. We also introduce a Multi-Modal Feature Alignment (MMFA) loss, encouraging the features of adversarial outputs to align with those of the desired outputs at the semantic level. Beyond bypassing, we present defensive LoRA patching, embedding visible warnings in the outputs as a complementary solution to mitigate this newly identified security vulnerability. With only 1,000 facial examples and a single epoch of fine-tuning, LoRA patching successfully defeats multiple proactive defenses. These results reveal a critical weakness in current paradigms and underscore the need for more robust Deepfake defense strategies. Our code is available at https://github.com/ZOMIN28/LoRA-Patching

Most research efforts in the multimedia forensics domain have focused on detecting forgery audio-visual content and reached sound achievements. However, these works only consider deepfake detection as a classification task and ignore the case where partial segments of the video are tampered with. Temporal forgery localization (TFL) of small fake audio-visual clips embedded in real videos is still challenging and more in line with realistic application scenarios. To resolve this issue, we propose a universal context-aware contrastive learning framework (UniCa-CLF) for TFL. Our approach leverages supervised contrastive learning to discover and identify forged instants by means of anomaly detection, allowing for the precise localization of temporal forged segments. To this end, we propose a specialized context-aware perception layer that utilizes a heterogeneous activation operation and an adaptive context updater to construct a context-aware contrastive objective, which enhances the discriminability of forged instant features by contrasting them with genuine instant features in terms of their distances to the global context. An efficient context-aware contrastive coding is introduced to further push the limit of instant feature distinguishability between genuine and forged instants in a supervised sample-by-sample manner, suppressing the cross-sample influence to improve temporal forgery localization performance. Extensive experimental results over five public datasets demonstrate that our proposed UniCaCLF significantly outperforms the state-of-the-art competing algorithms. The source code and pre-trained models of our proposed UniCaCLF are made publicly available at GitHub repository.TimeTimeTimeTime

Moving Object Detection (MOD) can be outsourced to the cloud for computational convenience, in which case the video must be encrypted to protect privacy. Secure video MOD methods designed to perform MOD on encrypted video are still in their infancy. In this paper, we present an attention-based framework for privacy-preserving MOD in compressed videos. On the user side, we adopt selective video encryption for the compressed video, while in the cloud, we extract the Compressed Video entropy-coded Syntax Elements (CVSE) from the encrypted video. Since the extracted CVSE data lacks sufficient motion information and contains noise, we introduce a two-stage training process. In the first phase, we propose a new deep learning-based approach for interpolating CVSE-based motion feature maps, addressing a significant drawback of traditional methods that rely exclusively on empirical interpolation algorithms. In the second stage, we propose a specialized backbone tailored for feature extraction from sparse CVSE data. We then design an attention-based neck that focuses on areas with denser motion and varying sizes of moving objects. Experimental results on two public datasets, VIRAT and DUKE-MTMC, show that our framework achieves state-of-the-art detection performance. Compared to previous secure solutions, the proposed method exhibits more robustness in challenging scenes.

Talking face generation (TFG) allows for producing lifelike talking videos of any character using only facial images and accompanying text. Abuse of this technology could pose significant risks to society, creating the urgent need for research into corresponding detection methods. However, research in this field has been hindered by the lack of public datasets. In this paper, we construct the first large-scale multi-scenario talking face dataset (MSTF), which contains 22 audio and video forgery techniques, filling the gap of datasets in this field. The dataset covers 11 generation scenarios and more than 20 semantic scenarios, closer to the practical application scenario of TFG. Besides, we also propose a TFG detection framework, which leverages the analysis of both global and local coherence in the multimodal content of TFG videos. Therefore, a region-focused smoothness detection module (RSFDM) and a discrepancy capture-time frame aggregation module (DCTAM) are introduced to evaluate the global temporal coherence of TFG videos, aggregating multi-grained spatial information. Additionally, a visual-audio fusion module (V-AFM) is designed to evaluate audiovisual coherence within a localized temporal perspective. Comprehensive experiments demonstrate the reasonableness and challenges of our datasets, while also indicating the superiority of our proposed method compared to the state-of-the-art deepfake detection approaches.

Image forgery localization (IFL) is a crucial technique for preventing tampered image misuse and protecting social safety. However, due to the rapid development of image tampering technologies, extracting more comprehensive and accurate forgery clues remains an urgent challenge. To address these challenges, we introduce a novel information-theoretic IFL framework named SUMI-IFL that imposes sufficiency-view and minimality-view constraints on forgery feature representation. First, grounded in the theoretical analysis of mutual information, the sufficiency-view constraint is enforced on the feature extraction network to ensure that the latent forgery feature contains comprehensive forgery clues. Considering that forgery clues obtained from a single aspect alone may be incomplete, we construct the latent forgery feature by integrating several orthogonal individual image features. Second, based on the information bottleneck, the minimality-view constraint is imposed on the feature reasoning network to achieve an accurate and concise forgery feature representation that counters the interference of task-unrelated features. Extensive experiments show the superior performance of SUMI-IFL to existing state-of-the-art methods, not only on in-dataset comparisons but also on cross-dataset comparisons.

Multimodal fake news detection aims to automatically identify real or fake news, thereby mitigating the adverse effects caused by such misinformation. Although prevailing approaches have demonstrated their effectiveness, challenges persist in cross-modal feature fusion and refinement for classification. To address this, we present a residual-aware compensation network with multi-granularity constraints (RaCMC) for fake news detection, that aims to sufficiently interact and fuse cross-modal features while amplifying the differences between real and fake news. First, a multiscale residual-aware compensation module is designed to interact and fuse features at different scales, and ensure both the consistency and exclusivity of feature interaction, thus acquiring high-quality features. Second, a multi-granularity constraints module is implemented to limit the distribution of both the news overall and the image-text pairs within the news, thus amplifying the differences between real and fake news at the news and feature levels. Finally, a dominant feature fusion reasoning module is developed to comprehensively evaluate news authenticity from the perspectives of both consistency and inconsistency. Experiments on three public datasets, including Weibo17, Politifact and GossipCop, reveal the superiority of the proposed method.

Audio temporal forgery localization (ATFL) aims to find the precise forgery regions of the partial spoof audio that is purposefully modified. Existing ATFL methods rely on training efficient networks using fine-grained annotations, which are obtained costly and challenging in real-world scenarios. To meet this challenge, in this paper, we propose a progressive audio-language co-learning network (LOCO) that adopts co-learning and self-supervision manners to prompt localization performance under weak supervision scenarios. Specifically, an audio-language co-learning module is first designed to capture forgery consensus features by aligning semantics from temporal and global perspectives. In this module, forgery-aware prompts are constructed by using utterance-level annotations together with learnable prompts, which can incorporate semantic priors into temporal content features dynamically. In addition, a forgery localization module is applied to produce forgery proposals based on fused forgery-class activation sequences. Finally, a progressive refinement strategy is introduced to generate pseudo frame-level labels and leverage supervised semantic contrastive learning to amplify the semantic distinction between real and fake content, thereby continuously optimizing forgery-aware features. Extensive experiments show that the proposed LOCO achieves SOTA performance on three public benchmarks.

With the development of generative artificial intelligence, new forgery methods are rapidly emerging. Social platforms are flooded with vast amounts of unlabeled synthetic data and authentic data, making it increasingly challenging to distinguish real from fake. Due to the lack of labels, existing supervised detection methods struggle to effectively address the detection of unknown deepfake methods. Moreover, in open world scenarios, the amount of unlabeled data greatly exceeds that of labeled data. Therefore, we define a new deepfake detection generalization task which focuses on how to achieve efficient detection of large amounts of unlabeled data based on limited labeled data to simulate a open world scenario. To solve the above mentioned task, we propose a novel Open-World Deepfake Detection Generalization Enhancement Training Strategy (OWG-DS) to improve the generalization ability of existing methods. Our approach aims to transfer deepfake detection knowledge from a small amount of labeled source domain data to large-scale unlabeled target domain data. Specifically, we introduce the Domain Distance Optimization (DDO) module to align different domain features by optimizing both inter-domain and intra-domain distances. Additionally, the Similarity-based Class Boundary Separation (SCBS) module is used to enhance the aggregation of similar samples to ensure clearer class boundaries, while an adversarial training mechanism is adopted to learn the domain-invariant features. Extensive experiments show that the proposed deepfake detection generalization enhancement training strategy excels in cross-method and cross-dataset scenarios, improving the model's generalization.

The success of face recognition (FR) systems has led to serious privacy concerns due to potential unauthorized surveillance and user tracking on social networks. Existing methods for enhancing privacy fail to generate natural face images that can protect facial privacy. In this paper, we propose diffusion-based adversarial identity manipulation (DiffAIM) to generate natural and highly transferable adversarial faces against malicious FR systems. To be specific, we manipulate facial identity within the low-dimensional latent space of a diffusion model. This involves iteratively injecting gradient-based adversarial identity guidance during the reverse diffusion process, progressively steering the generation toward the desired adversarial faces. The guidance is optimized for identity convergence towards a target while promoting semantic divergence from the source, facilitating effective impersonation while maintaining visual naturalness. We further incorporate structure-preserving regularization to preserve facial structure consistency during manipulation. Extensive experiments on both face verification and identification tasks demonstrate that compared with the state-of-the-art, DiffAIM achieves stronger black-box attack transferability while maintaining superior visual quality. We also demonstrate the effectiveness of the proposed approach for commercial FR APIs, including Face++ and Aliyun.

Current researches on Deepfake forensics often treat detection as a classification task or temporal forgery localization problem, which are usually restrictive, time-consuming, and challenging to scale for large datasets. To resolve these issues, we present a multimodal deviation perceiving framework for weakly-supervised temporal forgery localization (MDP), which aims to identify temporal partial forged segments using only video-level annotations. The MDP proposes a novel multimodal interaction mechanism (MI) and an extensible deviation perceiving loss to perceive multimodal deviation, which achieves the refined start and end timestamps localization of forged segments. Specifically, MI introduces a temporal property preserving cross-modal attention to measure the relevance between the visual and audio modalities in the probabilistic embedding space. It could identify the inter-modality deviation and construct comprehensive video features for temporal forgery localization. To explore further temporal deviation for weakly-supervised learning, an extensible deviation perceiving loss has been proposed, aiming at enlarging the deviation of adjacent segments of the forged samples and reducing that of genuine samples. Extensive experiments demonstrate the effectiveness of the proposed framework and achieve comparable results to fully-supervised approaches in several evaluation metrics.

Existing methods for reversible data hiding in ciphertext binary images only involve one data hider to perform data embedding. When the data hider is attacked, the original binary image cannot be perfectly reconstructed. To this end, this letter proposes multi-party reversible data hiding in ciphertext binary images, where multiple data hiders are involved in data embedding. In this solution, we use visual cryptography technology to encrypt a binary image into multiple ciphertext binary images, and transmit the ciphertext binary images to different data hiders. Each data hider can embed data into a ciphertext binary image and generate a marked ciphertext binary image. The original binary image is perfectly reconstructed by collecting a portion of marked ciphertext binary images from the unattacked data hiders. Compared with existing solutions, the proposed solution enhances the recoverability of the original binary image. Besides, the proposed solution maintains a stable embedding capacity for different categories of images.

An efficient robust watermarking method should be resistant to various distortions, including distortions from image processing and geometric attacks. Geometric attacks are significant challenges for watermarking methods because they destroy the synchronization of the watermark between the embedding side and extracting side. It is a considerable challenge to accomplish watermark synchronization for watermarking methods. To address this challenge, a novel robust watermarking method with synchronization is proposed. At the embedding side, the watermark and the template are embedded to generate the watermarked image. If the watermarked image is attacked, the watermark and template are also distorted. At the extracting side, a template enhanced-extracted network is proposed to achieve watermark synchronization. The template enhanced-extracted network effectively extracts the distorted template from the distorted image. The template-enhanced subnet can indirectly enhance the strength of the distorted template in the distorted image and improve the accuracy of the template-extracted subnet. The visual quality of the watermarked image is guaranteed because there is no need to embed the template with high strength. Then, the attack factor is predicted based on the distorted template. By leveraging this prediction, correct watermark extraction with synchronization is achieved. The experimental results demonstrate that the proposed watermarking method with synchronization yields excellent robustness under image processing, geometric attacks and combined attacks.

Deep neural network (DNN) models are widely used in various fields, such as pattern recognition and natural language processing, and provide considerable commercial value to their owners. Embedding a digital watermark in the model allows the legitimate owner to detect unauthorized use of the model. However, the existing DNN watermarking methods are vulnerable to model extraction attacks since the watermark task and the original model task are independent. In this article, a novel collaborative DNN watermarking framework is proposed to defend against model extraction attacks by establishing cooperation between the watermark generation and embedding. Specifically, the trigger samples are not only imperceptible to ensure perceptual stealth security but also infused with target-label information to guide the following feature associations. In the process of watermark embedding, the feature representation of trigger samples is forced to be similar to that of the task distribution samples via feature coupling. Consequently, the trigger samples from our framework can be recognized in the stolen model as task distribution samples, so that the ownership of the model can be successfully verified. Extensive experiments on CIFAR10, CIFAR100, and ImageNet demonstrate the effectiveness and superior performance of the proposed watermarking framework against various model extraction attacks.

This paper proposes a novel method of vacatingroom-after-encryption reversible data hiding in encrypted image (VRAE RDHEI), which uses the ideas of channel modeling and channel coding to achieve the enhancement of capacity. The framework of the proposed method maps the processes of data embedding and image recovery to a virtual noisy channel for transmitting special flag bits of image content, and then it uses the systematic polar code to ensure error-free transmission for reversible data hiding. On the data hider side, to reversibly vacate room for secret data, the selected bits of the encrypted image are transformed to flag bits and then encoded to fewer parity bits by systematic polar code. On the receiver side, the secret data can be extracted without error and separate from image recovery. To recover the image, the receiver uses pixel prediction to obtain the noisy flag bits and decodes them to the original flag bits by a special channel knowledge-based decoding algorithm with the parity bits. Then, the original image can be recovered by the flag bits. The experimental results prove that the proposed method outperforms the state-of-the-art VRAE methods.

Deep Learning image forgery localization methods have achieved remarkable results but cannot maintain comparable performance when the forgery images are JPEG compressed, a format that is widely used in daily information transmission. The robustness against JPEG compression has become a bottleneck to the practical application of image forgery localization. To address this issue, a robust image forgery localization framework is proposed against the performance degradation caused by JPEG compression. Specifically, a cutting-edge progressive disentanglement strategy is proposed that incorporates coarse-grained image disentanglement to mitigate the detrimental effects of general JPEG compression, while harnessing the ability of fine-grained element disentanglement to separate multi-scale artifacts, thereby minimizing interference from content information. Moreover, the decision strategy is carefully designed to reinforce subtle signals from tampered areas, including artifacts fusion block reasoning multi-scale artifacts and dual attention block that learn more about forgery-related features. Extensive visualizations and experiments demonstrate that our method can achieve competitive performance in general JPEG-resistant image forgery localization, especially in the performance of generalization experiments.

As deepfake technology advances, forgery detection techniques have evolved beyond simple classification to include fine-grained localization. However, existing deepfake localization methods struggle with with real-world deepfake videos, which are often multi-face scenarios with only some parts manipulated. To address the above-mentioned problems, we propose a Multi-View Inconsistency Measurement (MVIM) network that simultaneously measures inconsistencies from noise and temporal view to detect and locate tampered regions. Specifically, considering the noise inconsistencies in multi-face scenarios where fake faces have inconsistent noise patterns compared to real faces and backgrounds, we design a Noise Inconsistency Measurement (Noise-IM) module that measures noise similarity among faces and between faces and backgrounds using a masked attention mechanism to identify suspected tampered regions in noise domain. Since facial jitter of tampered regions in deepfake videos is observed to be more intense than that of real regions, we design a Temporal Inconsistency Measurement (Temporal-IM) module which adopts self-attention mechanism and fine-grained bi-direction convolutions to capture tampering traces between frames in temporal domain. Inconsistency features obtained by the two modules are fused for detecting and locating tampered regions. The superiority of our MVIM network is verified by extensive experiments with many state-of-the-art methods in different benchmark datasets.

In cloud computing, it is necessary to outsource image processing algorithms securely without exposing private image content. The scale-invariant feature transform (SIFT) is a famous local descriptor widely used in computer vision. There are already some privacy-preserving schemes for computing SIFT on encrypted images. However, the state-of-the-art works have to convert fixed-point numbers into their binary representations, which reduces efficiency and accuracy. In this paper, we propose a novel privacy-preserving SIFT scheme built from secure protocols designed explicitly for fixed-point numbers to solve this problem. Specifically, using RLWE-based homomorphic encryption, we propose word-wise protocols to perform secure division, square root operation, comparison, derivation, and matrix inversion in a single-instruction multiple-data manner. These protocols allow direct processing of fixed-point numbers without converting them to binary numbers, thus achieving high computational efficiency. We have also realized critical SIFT steps missing from previous works, including Euclidean gradient amplitude computation, histogram peak interpolation, and precise interval localization, leading to improved accuracy of SIFT features in the encrypted domain. We conduct security analysis and perform extensive experiments to evaluate the execution efficiency and accuracy. The experimental results show that the proposed scheme outperforms the state-of-the-art works in terms of computational efficiency and accuracy.

Generative steganography stands as a promising technique for information hiding, primarily due to its remarkable resistance to steganalysis detection. Despite its potential, hiding a secret image using existing generative steganographic models remains a challenge, especially in lossy or noisy communication channels. This paper proposes a robust generative steganography model for hiding full-size image. It lies on three reversible concatenated mappings proposed. The first mapping uses VQGAN with an order-preserving codebook to compress an image into a more concise representation. The second mapping incorporates error correction to further convert the representation into a robust binary representation. The third mapping devises a distribution-preserving sampling mapping that transforms the binary representation into the latent representation. This latent representation is then used as input for a text-to-image Diffusion model, which generates the final stego image. Experimental results show that our proposed scheme can freely customize the stego image content. Moreover, it simultaneously attains high stego and recovery image quality, high robustness, and provable security.

Generative steganography is renowned for its exceptional undetectability. However, prevalent generative methods often have insufficient capacity for concealing secret images. Furthermore, the sensitivity of commonly utilized generative models exacerbates the challenge of ensuring robustness against channel distortions such as JPEG compression. In this paper, we introduce a generative image hiding network that employs two invertible generators to transform secret images into stego images within a disparate image domain. Additionally, we seamlessly integrate an up-and-down sampling module (UDM) within these generators to facilitate efficient decoupling of the intermediate representations obtained by each generator. The UDM serves multiple purposes: preserving coherence between the intermediate representations, enhancing resilience against JPEG compression, and safeguarding the confidentiality of the concealed images. To address the complexity of mapping both uncompressed and compressed stego images to a unified intermediary representation, we implement two distinct flows for the forward and backward processes of the generator associated with the stego images. The experimental results show that our scheme offers concurrent advantages in terms of full-size image hiding ability, undetectability, confidentiality, and robustness.

The evolution of advanced artificial intelligence generated content approaches has heightened concerns about deepfake, due to the sophisticated forgeries and concealed appearances they produce. To this end, the pre-trained Vision Transformer (ViT) model has become a de facto choice for deepfake detection, thanks to its powerful learning capability. Despite favorable results achieved by existing ViT-based methods, they have inherent limitations that could result in suboptimal performance in scenarios with continuously evolving forgery techniques, such as overfitting to single forgery patterns or placing excessive emphasis on dominant forgery regions. In this paper, we propose CUTA, a simple yet effective deepfake detection paradigm that utilizes ViT adapters as the medium and fully exploits the spatial- and frequency-domain features of given images to overcome the limitations of existing methods. Specifically, CUTA focuses onfrequency domain maskingwithin the input space, which obscures parts of the high-frequency image to intensify the training challenge while preserving subtle forgery cues in the frequency domain to facilitate comprehensive forgery representations. Furthermore, we propose two task-customized modules within the ViT model, i.e., thetexture enhancement moduleand themulti-scale perceptron module, to seamlessly integrate local texture and rich contextual features. These two modules ensure an organic interaction between the task-specific forgery patterns and general semantic features within the pre-trained ViT framework. The experimental results on several publicly available benchmark datasets demonstrate CUTA’s superiority in performance, particularly showcasing its significant advantages in both cross-dataset and cross-manipulation scenarios.

The purpose of robust image watermarking is to embed a watermark into a carrier image in an invisible form and extract the watermark successfully even under noise interference conditions to achieve copyright confirmation and traceability. Although watermarking methods based on deep learning can improve the robustness by adding a noise simulation layer, few theoretical analyses of the codec structure have been conducted. Theoretical explainability is the theoretical basis for developing a network architecture, which plays a guiding role in network development. On the basis of the interpretability of convolutional networks, this paper analyzes the mathematical process of embedding and extracting watermarks in codecs and proposes a novel watermarking framework based on multi-layer watermark feature fusion. Specifically, the encoder can be a convolutional network structure of arbitrary depth, whereas the decoder needs only to adopt its corresponding deconvolution structure. To improve the quality and robustness of the generated watermarked image, the watermark is associated with an arbitrary layer feature space in the decoder. In the decoder, the network quickly converges to each original encoding feature space through the deconvolution structure, thus decoupling the watermark features. Finally, the watermark is extracted via the automatic fusion of multi-layer watermark features. The experimental results show that the proposed method is suitable for few-shot learning, and its invisibility, robustness and generalization performance on multiple datasets are significantly better than those of other advanced methods.

Deep learning solutions for privacy anonymization in video action recognition face significant computational challenges and difficulties in collecting privacy labels. To address these issues, we propose a fast privacy-preserving action recognition framework based on compressed video bitstreams. Our framework introduces two novel anonymization methods for video bitstreams, transforming video data directly within the compressed domain into anonymized compressed data to protect sensitive video information. The anonymized and compressed video data exhibits a compact data representation, enhancing transmission efficiency. We propose a technique to reconstruct anonymized compressed data in the spatial domain and utilize deep learning technology for high-accuracy action recognition. The experiment results demonstrate that our framework excels in balancing privacy protection performance with action recognition, showing a significant trade-off advantage. Our approach also achieves an operational efficiency more than fivefold greater than existing methods.

Recently, a novel form of audio partial forgery has posed challenges to its forensics, requiring advanced countermeasures to detect subtle forgery manipulations within long-duration audio. However, existing countermeasures still serve a classification purpose and fail to perform meaningful analysis of the start and end timestamps of partial forgery segments. To address this challenge, we introduce a novel coarse-to-fine proposal refinement framework (CFPRF) that incorporates a frame-level detection network (FDN) and a proposal refinement network (PRN) for audio temporal forgery detection and localization. Specifically, the FDN aims to mine informative inconsistency cues between real and fake frames to obtain discriminative features that are beneficial for roughly indicating forgery regions. The PRN is responsible for predicting confidence scores and regression offsets to refine the coarse-grained proposals derived from the FDN. To learn robust discriminative features, we devise a difference-aware feature learning (DAFL) module guided by contrastive representation learning to enlarge the sensitive differences between different frames induced by minor manipulations. We further design a boundary-aware feature enhancement (BAFE) module to capture the contextual information of multiple transition boundaries and guide the interaction between boundary information and temporal features via a cross-attention mechanism. Extensive experiments show that our CFPRF achieves state-of-the-art performance on various datasets, including LAV-DF, ASVS2019PS, and HAD.

Capturing displayed images using portable cameras has become familiar among multimedia pirates, necessitating the urgent requirement of camera-shooting resilient watermarking schemes. In this paper, we consider the stealers who only record parts of images, and propose a robust watermarking scheme at the image instance level. This scheme consists of an encoding end, a noise layer, and a decoding end. The encoding end first selects specific watermarking regions associated with segmented image instances. Afterwards, an encoder is employed to embed watermark sequences into the RGB color model of these watermarking regions. At last, templates are embedded to product the final watermarked images. Specifically, our suggested template-based resynchronization comprises a template embedding module at the encoding end and a geometric correction module at the decoding end. The former embeds templates by a correlation-aware multiplicative spread spectrum with an adaptive amplitude, while the latter learns a calibrator to estimate the perspective projection. Experiments on both simulation and real-world scenarios support that the proposed scheme effectively resists camera-shooting attacks with various shooting conditions, regardless of whether the entire displayed images have been captured.

The risk of malicious exploitation of advanced image steganography necessitates the removal of hidden information from images. However, it is crucial to preserve the visual quality of the images undergoing processed. This paper suggests a geometrical attack in frequency domain (GAF) to address this challenge. GAF employs a thin plate spline (TPS) to slightly geometrically perturb the frequency components of the stego image. It incorporates a channel weight estimator and a frequency jammer. The channel weight estimator assigns perturbation strengths to each DCT channel, while the frequency jammer performs the TPS transform on the DCT channels using the assigned perturbation strengths. Experimental results demonstrate that the proposed approach effectively hinders secret image recovery with a little distortion to the stego images. Furthermore, it well preserves the visual quality of clear images that do not contain secret information.

The misuse of Deepfakes to create unauthorized fake facial images and videos poses a growing threat to personal privacy and social stability. Proactive defense algorithms have been proposed to prevent this fraud by injecting adversarial perturbations into facial images. However, these perturbations are sensitive to the lossy compression on online social networks (OSNs). Recent studies have attempted to produce compression resistance by modeling compression at the pixel level. However, accurate modeling is challenging due to the customization of proprietary compression mechanisms by different OSNs. In this paper, we propose a Robust Adversarial Perturbation (DF-RAP) that provides persistent protection for facial images under OSN compression. Specifically, a novel Compression Approximation GAN (ComGAN) is designed to explicitly model OSN compression. The well-trained ComGAN is then incorporated as a sub-module of the target Deepfake model to derive DF-RAP. Furthermore, we reveal a commonality among various OSNs, i.e., that the lossy compression employed tends to destroy perturbations. Based on this, a novel objective-level destruction-aware constraint (DAC) is introduced during ComGAN training. The extensive experimental results show that DF-RAP can effectively protect facial images from Deepfakes under complex OSN compression, especially for OSNs employing more stringent compression. We also investigate the lossy operation mechanisms employed by widely used OSN platforms and build an OSN-transmission dataset based on the CelebA to facilitate future research.

In recent years, audio spoofing detection has received widespread attention for protecting personal privacy and social security. Despite the significant progress achieved in audio single-view spoofing detection, challenges remain with regard to addressing unknown spoofing attacks in realistic scenarios. To solve these challenging problems, in this paper, we introduce a novel audio multi-view spoofing detection framework (AMSDF), whose goal is to capture both intra-view and inter-view cues by measuring correlations within audio multi-view features (i.e., audio-emotion-text) for audio spoofing detection. In general, different view features are inherently interconnected in the real patterns, while they may present unnatural correlations in the spoofing patterns. Therefore, more discriminative cues can be mined by utilizing their complex interactions, which is beneficial to the audio spoofing detection task. To this end, an intra-view graph attention mechanism (IGAM) is first utilized to aggregate each intra-view node within the same view. Subsequently, a heterogeneous graph fusion module (HGFM) is applied to measure correlations within inter-view nodes, which are enhanced with a master node for comprehensive analysis purposes. Finally, a group-based readout scheme (GRS) is designed to capture and preserve the most distinctive cues by leveraging the strengths of different feature sets, thereby effectively distinguishing subtle differences between real and spoofing audio. The experimental results show that our proposed framework can achieve better performance than that of the state-of-the-art methods, especially in realistic scenarios. The code and pre-trained models are available athttps://github.com/ItzJuny/AMSDF.

With the rapid progress of deepfake techniques in recent years, facial video forgery can generate highly deceptive video content and bring severe security threats. And detection of such forgery videos is much more urgent and challenging. Most existing detection methods treat the problem as a vanilla binary classification problem. In this article, the problem is treated as a special fine-grained classification problem since the differences between fake and real faces are very subtle. It is observed that most existing face forgery methods left some common artifacts in the spatial domain and time domain, including generative defects in the spatial domain and interframe inconsistencies in the time domain. And a spatial-temporal model is proposed which has two components for capturing spatial and temporal forgery traces from a global perspective, respectively. The two components are designed using a novel long-distance attention mechanism. One component of the spatial domain is used to capture artifacts in a single frame, and the other component of the time domain is used to capture artifacts in consecutive frames. They generate attention maps in the form of patches. The attention method has a broader vision which contributes to better assembling global information and extracting local statistic information. Finally, the attention maps are used to guide the network to focus on pivotal parts of the face, just like other fine-grained classification methods. The experimental results on different public datasets demonstrate that the proposed method achieves state-of-the-art performance, and the proposed long-distance attention method can effectively capture pivotal parts for face forgery.

Abstract In this paper, a geometrically invariant image watermarking scheme is proposed by exploiting multilevel histogram shapes. The embedding procedure starts by decomposing the host image with the first level Haar wavelet. After that, histograms are extracted from the approximation subband via several rounds, which are used to embed watermark bits. Each round of embedding first extracts a histogram at a specified level. Then the histogram is split into fragments, into which a number of watermark bits can be embedded. In this way, a considerable watermarking capacity is available. Besides, a histogram adjustment in the first embedding round is suggested to guarantee good population of histogram bins. Experimental results support its robustness against various common attacks and geometric attacks. Moreover, the scheme can embed multiple watermark sequences with various robustness and capacity profiles, which enriches its practical applications.

In order to reply the potential security issues caused by the tampering of digital images, many image forensics approaches based on deep learning have been proposed in recent years. However, the interpretability of deep learning-based approaches has not been fully considered. In this paper, an interpretable image tampering detection approach is proposed. It consists of suspicious tampered region detection (STRD) module and cooperative game module. The STRD module, inspired by YOLO, combines the shallow-level and deep-level features to discriminate different tampering types of suspicious tampered regions in complex scenes, and also performs well in detecting small tampered regions. The prediction of STRD module could be extended to suspicious box and the payoff of cooperative game module. The cooperative game module utilizes the Shapley interaction index as the strategy to measure the information gain of image pixels. The Shapley interaction index disentangles the multi-order interaction between the pixels of the image, and we discover that image tampering mainly affects low-order interaction of the image. The final detection result is obtained by combining the pixels that contribute greatly to the payoff with the suspicious box. The proposed approach provides a new thought for the interpretability of image forensics and could be broadly applied to other digital image forensic approaches. Extensive experimental results have demonstrated the proposed approach outperforms SOTA approaches, which also has good interpretability and robustness.

Image steganographic methods based on encoder-decoder model with end-to-end network architecture recently have been proposed. However, in steganographic applications, the feature map (called stego matrix) generated by the encoder needs to be rounded as a real stego image for the receiver. The loss of precision by rounding stego matrix leads to the decline in the accuracy of extracted secret messages. The challenge of using end-to-end network to preserve robustness against rounding operation is that it is non-differentiable. In this paper, we propose an anti-rounding image steganography method with separable fine-tuning network architecture which includes the joint training stage (JT-stage) and the separable fine-tuning stage (SF-stage). Firstly, in JT-stage, an embedded generator and a stego matrix extractor are jointly learned without rounding operation. Utilizing concatenation in embedded generator can realistically fuse cover image and secret messages. And the multi-scale fusion block and residual dense block in stego matrix extractor can make secret messages more correctly decoded. Moreover, the discriminator is constructed by generative adversarial nets (GAN) in JT-stage to effectively improve the authenticity and steganalysis security. Then, in SF-stage, the embedded generator is frozen, and the stego matrix is obtained and rounded as a stego image. A stego image extractor is constructed by fine-tuning the layers of the stego matrix extractor to improve the accuracy of message extraction. As the loss will not backpropagate in the embedded generator, the non-differentiability of rounding operation can be offset. Experiments show that the proposed separation fine-tuning network is robust to rounding operation, and effectively reduces the degradation of the image quality and steganalysis performance.

Robust steganography aims to hide message in cover data with high security and guarantee the success of its message extraction although it is disturbed in transmission channel. In this paper we propose a framework of coding scheme extended from Dual-Syndrome Trellis Codes (Dual-STCs) for robust adaptive steganography. We use the conditional probability distribution of correct stego bits conditioned on disturbed stego data as channel knowledge, and formulate error-correcting as maximizing this probability. By extending Dual-STCs to double-layered embedding, we design an iteratively decoding scheme for error-correcting two layer stego bits from their joint conditional probabilities, and strictly prove its convergence. Besides, we design a method to estimate these probability distributions from stego data pairs uploaded/downloaded from the lossy transmission channel. The channel knowledge can also be used by steganographer, and we propose a universal method to revise steganographic distortion values for higher robustness under the guidance of the channel knowledge. Compared with existing coding methods for robust steganography, our method can make use of channel knowledge to improve error correcting ability and meanwhile maintain high security, which is demonstrated by experimental results.

With the rapid development of face forgery techniques, the existing frame-based deepfake video detection methods have fell into a dilemma that frame-based methods may fail when encountering extremely realistic images. To overcome the above problem, many approaches attempted to model the spatio-temporal inconsistency of videos to distinguish real and fake videos. However, current works model spatio-temporal inconsistency by combining intra-frame and inter-frame information, but ignore the disturbance caused by facial motions that would limit further improvement in detection performance. To address this issue, we investigate into long and short range inter-frame motions and propose a novel dynamic difference learning method to distinguish between the inter-frame differences caused by face manipulation and the inter-frame differences caused by facial motions in order to model precise spatio-temporal inconsistency for deepfake video detection. Moreover, we elaborately design a dynamic fine-grained difference capture module (DFDC-module) and a multi-scale spatio-temporal aggregation module (MSA-module) to collaboratively model spatio-temporal inconsistency. Specifically, the DFDC-module applies self-attention mechanism and fine-grained denoising operation to eliminate the differences caused by facial motions and generates long range difference attention maps. The MSA-module is devised to aggregate multi-direction and multi-scale temporal information to model spatio-temporal inconsistency. The existing 2D CNNs can be extended into dynamic spatio-temporal inconsistency capture networks by integrating the proposed two modules. Extensive experimental results demonstrate that our proposed algorithm steadily outperforms state-of-the-art methods by a clear margin in different benchmark datasets.

Document images often contain essential and sensitive information. With image editing software, one can easily manipulate the semantic meaning of the document image by copy-move, splicing, and removal, which causes many security issues. Hence, the document image forgery detection and localization are of great significance. In this paper, a novel two-stream network is proposed to detect and locate the forgery regions of document images. One stream captures forgery traces from the spatial information, including unnatural boundaries, contrast/brightness inconsistencies, and blurring. The other stream eliminates the image semantics by six residual filters, then capture the anomalous features between neighboring pixels introduced by upsampling, downsamping, and inpainting. Finally, the discriminant network is introduced to deeply fuse features from the two streams, and discriminates whether each image patch is forged or not using global average pooling and fully connected layer. Extensive experimental results on our constructed document image data set and the Security AI Challenger Program data set demonstrate the proposed two-stream network outperforms state-of-the-art methods. This proposed two-stream network is also better than each single stream, and is robust to some common forgery postprocessing operations.

Syndrome-trellis codes (STCs) are commonly used in image steganographic schemes, which aim at minimizing the embedding distortion, but most distortion models cannot capture the mutual interaction of embedding modifications (MIEMs). In this article, a secure halftone image steganographic scheme based on a feature space and layer embedding is proposed. First, a feature space is constructed by a characterization method that is designed based on the statistics of 4 ×4 pixel blocks in halftone images. Upon the feature space, a generalized steganalyzer with good classification ability is proposed, which is used to measure the embedding distortion. As a result, a distortion model based on a hybrid feature space is constructed, which outperforms some state-of-the-art models. Then, as the distortion model is established on the statistics of local regions, a layer embedding strategy is proposed to reduce MIEM. It divides the host image into multiple layers according to their relative positions in 4 ×4 blocks, and the embedding procedure is executed layer by layer. In each layer, any two pixels are located at different 4 ×4 blocks in the original image, and the distortion model makes sure that the calculation of pixel distortions is independent. Between layers, the pixel distortions of the current layer are updated according to the previous embedding modifications, thus reducing the total embedding distortion. Comparisons with prior schemes demonstrate that the proposed steganographic scheme achieves high statistical security when resisting the state-of-the-art steganalysis.

As one of the most important topics in image forensics, resampling detection has developed rapidly in recent years. However, the robustness to JPEG compression is still challenging for most classical spectrum-based methods, since JPEG compression severely degrades the image contents and introduces block artifacts in the boundary of the compression grid. In this article, we propose a method to estimate the upscaling factors on double JPEG compressed images in the presence of image upscaling between the two compressions. We first analyze the spectrum of scaled images and give an overall formulation of how the scaling factors along with the parameters of JPEG compression and image contents influence the appearance of tampering artifacts. The expected positions of five kinds of characteristic peaks are analytically derived. Then, we analyze the features of double JPEG compressed images in the block discrete cosine transform (BDCT) domain and present an inverse scaling strategy for the upscaling factor estimation with a detailed proof. Finally, a fusion method is proposed that through frequency-domain analysis, a candidate set of upscaling factors is given, and through analysis in the BDCT domain, the optimal estimation from all candidates is determined. The experimental results demonstrate that the proposed method outperforms other state-of-the-art methods.

The existing models of reversible data hiding in encrypted images (RDH-EI) are based on single data-hider, where the original image cannot be reconstructed when the data-hider is damaged. To address this issue, this article proposes a novel model with multiple data-hiders for RDH-EI based on secret sharing. It divides the original image into multiple different encrypted images with the same size of the original image and distributes them to multiple different data-hiders for data hiding. Each data-hider can independently embed data into the encrypted image to obtain the corresponding marked encrypted image. The original image can be losslessly recovered by collecting sufficient marked encrypted images from undamaged data-hiders when individual data-hiders are subjected to potential damage. This further protects the security of the original image. We provide four cases of the proposed model, namely, two joint cases and two separable cases. From the proposed model, we derive a separable RDH-EI method with high-capacity. Experimental results are presented to illustrate the effectiveness of the proposed method.

The rapid development of DeepFake technologies has brought great challenges to the authenticity of video contents. It is of vital importance to develop DeepFake detection methods, among which three-dimensional (3D) convolution neural networks (CNN) have attracted wide interest and achieved satisfying performances. However, there are few 3D CNNs designed for DeepFake detection and the parameters of them are large, which cause heavy memory and storage consumption. In this paper, a lightweight 3D CNN is proposed for DeepFake detection. Channel transformation module is designed to extract features with much fewer parameters in higher level. Serving as spatial-temporal module, 3D CNNs are adopted to fuse the spatial features in time dimension. To suppress frame content and highlight frame texture, spatial rich model features are extracted from the input frames, which helps the spatial-temporal module achieve better performance. Experimental results show that the number of parameters of the proposed network is much less than those of other networks and the proposed network outperforms other state-of-the-art DeepFake detection methods on mainstream DeepFake data sets.

Social networks are everywhere and currently transmitting very large messages. As a result, transmitting secret messages in such an environment is worth researching. However, the images used in transmitting messages are usually compressed with a JPEG compression channel, which is lossy and damages the transmitted data. Therefore, to prevent secret messages from being damaged, a robust JPEG steganography is urgently needed. In this paper, a secure robust JPEG steganographic scheme based on an autoencoder with an adaptive BCH encoding (Bose-Chaudhuri-Hocquenghem encoding) is proposed. In particular, the autoencoder is first pretrained to fit the transformation relationship between the JPEG image before and after compression by the compression channel. In addition, the BCH encoding is adaptively utilized according to the content of cover image to decrease the error rate of secret message extraction. The DCT (Discrete Cosine Transformation) coefficient adjustment based on practical JPEG channel characteristics further improves the robustness and statistical security. Comparisons with prior state-of-the-art schemes demonstrate that the proposed robust JPEG steganographic algorithm can provide a more robust performance and statistical security.

In many reversible data hiding (RDH) methods for halftone images, the traditional embedding process embeds a 1-bit secret message into each embeddable pixel or pattern. To improve the embedding efficiency and payload, we propose an RDH method used in halftone images based on the dynamic embedding states group (DESG), which can embed at least 1 bit of secret messages per embeddable pixel or pattern. First, by exploiting the statistical features of$4 \times 4$patterns and the state sequences in each image, the DESG is constructed dynamically, including$n$embedding states with their state patterns and state sequences. Then, secret messages are encoded by matching the longest common subsequence according to the DESG, which are split into several state sequences. The state sequences are embedded by Markov transitions between these$n$changing state patterns. Finally, reversibility is achieved by recording the DESG as the overhead information in RDH. Experiments show that the construction of DESG can improve the embedding efficiency under the same number of embeddable pixels or patterns, and the visual distortion is also significantly reduced by flipping fewer pixels.

Most state-of-the-art halftone image steganographic techniques only consider the flipping distortion according to the human visual system, which are not always secure when they are attacked by steganalyzers. In this paper, we propose a halftone image steganographic scheme that aims to generate stego images with good visual quality and strong statistical security of anti-steganalysis. First, the concept of pixel density is proposed and a novel construction called pixel density histogram (PDH) is proposed to design a “embedding” scheme for halftone images. Then, we optimize density pair selection to select density blocks that can improve visual quality. Finally, the messages are embedded through pixel density transition, where a novel pixel flipping strategy is proposed, which can maintain the structural dependence by optimizing the pixel mesh Markov transition matrix (PMMTM). The experimental results demonstrate that the proposed steganography scheme can achieve strong statistical security of anti-steganalysis with good visual quality without degrading the embedding capacity.

Resampling detection is one of the most important topics in image forensics, and the most widely used method in resampling detection is spectral analysis. Since JPEG is the most widely used image format, it is reasonable that the resampling operation is processed on JPEG images. JPEG block artifacts bring severe interference to spectrum-based methods and degrade the detection performance. In addition, the spectral characteristics of the downscaling scenarios are very weak. The detection of downscaling still presents a considerable challenge to forensic applications. In this paper, we propose a method to estimate the downscaling factors of pre-JPEG compressed images in the presence of image downscaling after JPEG compressions. We first analyze the spectrum of scaled images and give an exact formulation of how the scaling factors influence the appearance of periodic artifacts. The expected positions of the characteristic resampling peaks are analytically derived. For the downscaling scenario, the shifted JPEG block artifacts produce periodic peaks, which cause misdetection in the characteristic peak. We find that the interval between the adjacent extrema of difference images obeys the geometric distribution and the distribution has periodic peaks for JPEG images. Hence, we adopt the difference image extremum interval histogram and combine the spectral method to obtain the final estimation. The experimental results demonstrate that the proposed detection method outperforms some state-of-the-art methods.

Utilizing statistical models of binary images is a common and effective means to steganalyze binary images, and the design of the statistical model is essential to the performance of steganalysis. In this paper, we propose a new model based on a histogram of pixel structuring elements (SEs), which is a suitable representation of a binary image for the task of steganalysis. The texture property and the dependency among pixels are considered inside the SEs. The SEs with different patterns will be evaluated comprehensively according to a statistical criterion, and some of them will be selected to construct the feature set for training the steganalyzer. The distributions of these selected SEs, which contain many highly flippable pixels, will be emphasized by the criterion, and they can reflect the difference between cover images and stego-images. Finally, a series of experiments are conducted on two datasets, and the results show that the proposed scheme significantly outperforms state-of-the-art schemes.

In this paper, a binary image steganographic scheme is presented, which aims at minimizing the embedding distortions measured by prediction. A prediction model of the center pixel's value is established in a 3 × 3 local region. A concept of “uncertainty” is introduced to represent the prediction result and the uncertainty is defined as the proximity of probabilities about whether the center pixel is black or white. A pixel with high uncertainty means that it is hard to distinguish whether it has been flipped or not, and thus the distortion introduced by flipping this pixel is small. The uncertainty is an appended statistical explanation of human visual perception and the distortion measurement based on it can evaluate the embedding changes on both vision and statistics. Benefiting from the statistics, uncertainty can evaluate the distortion influence in an extended local region. To play the advantage of distortion measurement, the syndrome-trellis code (STC) is employed to minimize the embedding distortions. Comparisons with prior schemes demonstrate that the proposed steganographic scheme achieves high vision imperceptibility and statistical security.

Sample balancing includes sample selection and sample reweighting. Sample selection aims to remove some bad samples that may lead to bad local optima. Sample reweighting aims to assign optimal weights to samples to improve performance. In this article, we integrate a sample selection method based on self-paced learning into deep learning frameworks and study the influence of different sample selection strategies on training deep networks. In addition, most of the existing sample reweighting methods mainly take per-class sample number as a metric, which does not fully consider sample qualities. To improve the performance, we propose a novel metric based on the multiview semantic encoders to reweight the samples more appropriately. Then, we propose an optimization mechanism to embed sample weights into loss functions of deep networks, which can be trained in end-to-end manners. We conduct experiments on the CIFAR data set and the ImageNet data set. The experimental results demonstrate that our proposed sample balancing method can improve the performances of deep learning methods in several visual recognition tasks.

Some state-of-the-art binary image steganographic methods aim to generate stego images with good visual quality, while others focus more on the statistical security of the anti-steganalysis. This paper proposes a binary steganographic scheme that improves both of them by selecting more appropriate flipped pixels. First, a fused distortion measurement is developed that combines the advantages of flipping distortion measurement (FDM) and two data-carrying pixel location methods, including the edge adaptive grid method (EAG) and the “Connectivity Preserving” criterion (CPc). The FDM measures the distortion score by statistical features and achieves high-statistical security, while the EAG and CPc select pixels by analyzing the local texture structures based on visual quality. Then, to eliminate the interference brought by adjacent flipped pixels, a flipping position optimization strategy is proposed to find better positions for flipping pixels to further improve the steganographic performance. Experimental results have demonstrated that the proposed steganographic scheme can achieve stronger statistical security with better visual quality without degrading the embedding capacity.

Crowdsourcing systems which utilize the human intelligence to solve complex tasks have gained considerable interest and adoption in recent years. However, the majority of existing crowdsourcing systems rely on central servers, which are subject to the weaknesses of traditional trust-based model, such as single point of failure. They are also vulnerable to distributed denial of service (DDoS) and Sybil attacks due to malicious users involvement. In addition, high service fees from the crowdsourcing platform may hinder the development of crowdsourcing. How to address these potential issues has both research and substantial value. In this paper, we conceptualize a blockchain-based decentralized framework for crowdsourcing named CrowdBC, in which a requester's task can be solved by a crowd of workers without relying on any third trusted institution, users' privacy can be guaranteed and only low transaction fees are required. In particular, we introduce the architecture of our proposed framework, based on which we give a concrete scheme. We further implement a software prototype on Ethereum public test network with real-world dataset. Experiment results show the feasibility, usability, and scalability of our proposed crowdsourcing system.

Gait recognition is an important topic in biometrics. Current works primarily focus on recognizing a single person's walking gait. However, a person's gait will change when they walk with other people. How to recognize the gait of multiple people walking is still a challenging problem. This paper proposes an attribute discovery model in a max-margin framework to recognize a person based on gait while walking with multiple people. First, human graphlets are integrated into a tracking-by-detection method to obtain a person's complete silhouette. Then, stable and discriminative attributes are developed using a latent conditional random field (L-CRF) model. The model is trained in the latent structural support vector machine (SVM) framework, in which a new constraint is added to improve the multi-gait recognition performance. In the recognition process, the attribute set of each person is detected by inferring on the trained L-CRF model. Finally, attributes based on dense trajectories are extracted as the final gait features to complete the recognition. The experimental results demonstrate that the proposed method achieves better recognition performance than traditional gait recognition methods under the condition of multiple people walking together.

Solving Sylvester equation is a common algebraic problem in mathematics and control theory. Different from the traditional fixed-parameter recurrent neural networks, such as gradient-based recurrent neural networks or Zhang neural networks, a novel varying-parameter recurrent neural network, [called varying-parameter convergent-differential neural network (VP-CDNN)] is proposed in this paper for obtaining the online solution to the time-varying Sylvester equation. With time passing by, this kind of new varying-parameter neural network can achieve super-exponential performance. Computer simulation comparisons between the fixed-parameter neural networks and the proposed VP-CDNN via using different kinds of activation functions demonstrate that the proposed VP-CDNN has better convergence and robustness properties.

Learning deep representations have been applied in action recognition widely. However, there have been a few investigations on how to utilize the structural manifold information among different action videos to enhance the recognition accuracy and efficiency. In this paper, we propose to incorporate the manifold of training samples into deep learning, which is defined as deep manifold learning (DML). The proposed DML framework can be adapted to most existing deep networks to learn more discriminative features for action recognition. When applied to a convolutional neural network, DML embeds the previous convolutional layer's manifold into the next convolutional layer; thus, the discriminative capacity of the next layer can be promoted. We also apply the DML on a restricted Boltzmann machine, which can alleviate the overfitting problem. Experimental results on four standard action databases (i.e., UCF101, HMDB51, KTH, and UCF sports) show that the proposed method outperforms the state-of-the-art methods.

Most state-of-the-art binary image steganographic techniques only consider the flipping distortion according to the human visual system, which will be not secure when they are attacked by steganalyzers. In this paper, a binary image steganographic scheme that aims to minimize the embedding distortion on the texture is presented. We extract the complement, rotation, and mirroring-invariant local texture patterns (crmiLTPs) from the binary image first. The weighted sum of crmiLTP changes when flipping one pixel is then employed to measure the flipping distortion corresponding to that pixel. By testing on both simple binary images and the constructed image data set, we show that the proposed measurement can well describe the distortions on both visual quality and statistics. Based on the proposed measurement, a practical steganographic scheme is developed. The steganographic scheme generates the cover vector by dividing the scrambled image into superpixels. Thereafter, the syndrome-trellis code is employed to minimize the designed embedding distortion. Experimental results have demonstrated that the proposed steganographic scheme can achieve statistical security without degrading the image quality or the embedding capacity.