QCQi Cui

Qi Cui

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Qi Cui has a total of 35 co-authors. Here are the top co-authors by number of shared publications:

  • Zhili Zhou

    School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou, China

    7 shared publications
  • Ruohan Meng

    Nanjing University of Information Science and Technology

    6 shared publications
  • Zhangjie Fu

    Nanjing University of Information Science & Technology

    3 shared publications
  • Haoqin Yuan

    School of Cyber Science and Engineering, Nanjing University of Information Science and Technology, Nanjing, 210044, China; Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing, 210044, China

    2 shared publications
  • Qing Ming Jonathan Wu

    University of Windsor

    2 shared publications
  • Xingming Sun

    School of Computer and Software, Nanjing University of Information Science and Technology, Ning Liu Road, No. 219, Nanjing, 210044, China and Jiangsu Engineering Centre of Network Monitoring, Ning Liu Road, No. 219, Nanjing, 210044, China

    2 shared publications

Latest Publications

  • SDCA: Towards semantic-guided dual camouflage for deceiving human eyes and object detectors

    by Haoqin Yuan, Xianyi Chen, Qi Cui, Fazhan Liu, Zhangjie Fu

  • RAFS: Reversible Identity-Anonymization Face Swapping for Provenance Tracking

    by Jiancheng Li, Qi Cui, Peipeng Yu, Chip-Hong Chang, Zhangjie Fu

    Face-swapping technologies have rapidly emerged as a mainstream AI service across entertainment, social media, and virtual platforms. While current face-swapping methods offer highly realistic results, they also introduce significant privacy risks, as most current approaches require clear target faces, exposing users’ identities. Moreover, the absence of built-in authorization and forensic mechanisms renders these systems incapable of tracing or verifying manipulated content, raising critical issues over accountability and potential misuse. To address these challenges, we propose a privacy-preserving and forensics-enabled face-swapping framework that simultaneously safeguards user identity and enables robust post-hoc face recovery. Instead of relying on visible target faces, our method operates on non-facial target images, fundamentally preventing identity exposure at the source. To ensure provenance traceability, we embed the target face’s features into non-facial regions of the generated image via an imperceptible and reversible encoding scheme. To further enhance robustness, we introduce a mask-distortion simulation layer that bridges pre-/post-swap mask discrepancies and stabilizes face recovery under perturbations. Extensive experiments demonstrate that our method produces realistic face-swapped images without revealing the original identity, while enabling high-fidelity recovery under various adversarial conditions—validating its effectiveness in both privacy protection and forensic traceability.

  • Proactive Image Manipulation Detection and Tracing in Fake News

    by Ruohan Meng, Siyuan Yang, Zhili Zhou, Qi Cui, Kwok-Yan Lam, Zengwei Zheng, Alex C. Kot

    The pervasive spread of fake news, particularly through manipulated images, presents a consequential negative impact on society. To prevent fake news images from misleading the public, existing methods focus on verifying the authenticity of news images but ignore source traceability, leaving a gap in creating a complete forensic chain for reliable fake news detection. To simultaneously achieve the goals of authenticity verification and source tracing, we propose a proactive image tagging approach based on a design of Disentangled Invertible Neural Networks (DINN). It can simultaneously embed the dual-tags, i.e., authenticable tag and traceable tag, into each news image prior to publication, allowing for separate extraction for authenticity verification and source tracing. Within the proposed DINN, we design a parallel Feature Aware Projection Module (FAPM) to assist DINN in preserving essential tag information, thereby improving extraction accuracy. In addition, we introduce a Distance Metric-Guided Module (DMGM) that learns asymmetric one-class representations, enabling the dual-tags to exhibit different robustness performances under malicious manipulations. Extensive experiments on diverse datasets and unseen manipulations demonstrate that the proposed tagging approach achieves promising performances on both authenticity verification and source tracing for reliable fake news detection and outperforms the prior works.


  • FasterCReW: Performance or Efficiency? A Lightweight Conditional Residual DNN-Based Watermarking Based on FasterNet

    by Baowei Wang, Jianbo Zhang, Yufeng Wu, Qi Cui

    Deep neural networks (DNNs) based watermarking algorithms have made significant strides in recent years. However, existing methods either demand substantial resources for image feature extraction during watermark embedding, sacrificing efficiency, or completely neglect image texture information, resulting in suboptimal performance. Moreover, current algorithms struggle with real-time watermark extraction. To address these limitations, we propose a lightweight conditional residual watermarking (CReW) architecture. Specifically, CReW employs a Conditional Generative Adversarial Network (CGAN) framework to generate an adaptive residual image guided by the structure of the cover image, which is decoupled from the network to reduce computational complexity. This design enables CReW to achieve an optimal balance between performance and efficiency. Additionally, by directly optimizing the residual image to capture variations in watermark behavior under distortion, CReW significantly enhances robustness. Furthermore, we design redundancy coding blocks to increase the mutual information of the watermark, along with a patch-level discriminator to improve local patch discrimination, thereby further enhancing image quality. Finally, by reducing channel redundancy and leveraging FasterNet, we developed a low-complexity network architecture, Faster-CReW, which facilitates real-time watermark embedding and extraction. Extensive experimental results demonstrate that, despite having 36× fewer network parameters and 30× fewer floating point operations (FLOPs) than Adaptor, Faster-CReW exhibits excellent robustness against distortions such as cropout, JPEG compression, and Gaussian noise. Furthermore, Faster-CReW significantly outperforms other existing DNN-based watermarking algorithms in terms of running speed, achieving an 8× speed increase over UDH and a 28× increase over Adaptor on an Intel Core i7-8750H CPU.


  • AdvMap: Crafting Adversarial Maps to Counter AI Aimbot in First-Person Shooter Games

    by Kai Yan, Xianyi Chen, Qi Cui, Haoqin Yuan, Zhenshan Tan

    AI-based automatic aiming cheats (a.k.a., AI aimbots) have proliferated in first-person shooter (FPS) games, which grant malicious users an unfair gameplay advantage. Since AI aimbots operate independently of game data and are developed using object detection algorithms, they are difficult to detect with traditional anti-cheating methods. To actively counter AI aimbots, we propose AdvMap, which introduces invisible adversarial perturbations into game scene elements. In optimizing these adversarial perturbations, we design a mixture-of-misleading loss function that increases the total target confidence score within each misleading bounding box. It mitigates the risk of segment missing even when AdvMap is obscured, thereby enhancing the robustness. Besides, an L1-norm constraint with a small scale is employed during each update of the adversarial perturbations, which preserves the fidelity of the game scene. Additionally, to enable effective adaptation to interact with various elements within game environments, we introduce an Image-Subspace-Based Multi-Directional (ISMD) optimization strategy. It enables the adversarial perturbations to adaptively fit into each element by leveraging the mapping relationship between the game's 3D scenes and its corresponding 2D images. Furthermore, we construct a comprehensive benchmark, which includes various FPS games with different graphics styles and perspectives. Extensive experimental results demonstrate the efficacy of our method in countering various AI aimbot tools on different state-of-the-art object detection methods.


  • Meta Security Metric Learning for Secure Deep Image Hiding

    by Qi Cui, Weixuan Tang, Zhili Zhou, Ruohan Meng, Guoshun Nan, Yun-Qing Shi

    Deep Image Hiding (DIH) aims to imperceptibly hide images within image. To improve its security performance, some DIH methods design Security Metrics (SMs) to guide the learning of their hiding networks. However, these methods focus on optimizing their anti-steganalysis ability on specific SMs, resulting in inferior generalization ability. To overcome these limitations, in this paper, we introduce meta-learning into DIH and propose Meta Security Metric-based DIH (MSM-DIH). In the MSM-DIH, the Invertible Neural Network (INN)-based hiding network is learned under the guidance of a learnable meta SM generalized from multiple fixed source SMs, and each SM is composed of a metric network and a contrastive loss function. Specifically, MSM-DIH is trained with bi-level optimization. In the outer optimization, a meta SM is learned to assign higher security scores for more advanced stego images. Besides, the domain knowledge of steganalysis is transferred from the multiple pre-trained source metric networks to the meta metric network, so as to enhance the generalization ability of the meta SM. In the inner optimization, the hiding network is learned to generate more secure stego images according to the learned meta SM. Experimental results show that our MSM-DIH has achieved the best security performance in most cases.


  • ARES: On Adversarial Robustness Enhancement for Image Steganographic Cost Learning

    by Qi Cui, Zhili Zhou, Ruohan Meng, Shaowei Wang, Hongyang Yan, Q. M. Jonathan Wu

    Taking the steganalytic discriminators as the adversaries, the existing Generative Adversarial Networks (GAN)-based steganographic approaches learn the implicit cost functions to measure the embedding distortion for steganography. However, the steganalytic discriminators in these approaches are trained by the stego-samples with insufficient diversity, and their network structures offer very limited representational capacity. As a result, these steganalytic discriminators will not exhibit robustness to various steganographic patterns, which causes learning suboptimal cost functions, thus compromising the anti-steganalysis capability. To address this issue, we propose a novel GAN-based steganographic approach, in which the Diversified Inverse-Adversarial Training (DIAT) strategy and the Steganalytic Feature Attention (SteFA) structure are designed to train a robust steganalytic discriminator. Specifically, the DIAT strategy provides the steganalytic discriminator with an expanded feature space by generating diversified adversarial stego-samples; the SteFA structure enables the steganalytic discriminator to capture more various steganalytic features by employing the channel-attention mechanism on higher-order statistics. Consequently, the steganalytic discriminator can build a more precise decision boundary to make it more robust, which facilitates learning a superior steganographic cost function. Extensive experiments demonstrate that the proposed steganographic approach achieves promising anti-steganalysis capability over the state-of-the-arts under the same embedding payloads.


  • Encrypted Domain Secret Medical-Image Sharing With Secure Outsourcing Computation in IoT Environment

    by Jingwang Huang, Qi Cui, Zhili Zhou, Keping Yu, Ching-Nung Yang, Kim-Kwang Raymond Choo

    In existing secret medical-image sharing (SMIS) schemes, to protect and manage secret medical-images (SMIs), the sharing and recovery of each SMI are implemented by local servers of medical institutions. However, since a lot of SMIs are produced by personal smart terminal devices in Internet of Things (IoT) environment, directly implementing the sharing and recovery processes will cause excessive communication and computing burden for those local servers, which makes the existing SMIS schemes not suitable for IoT environment. To address the above issue, we propose an encrypted domain SMIS (Enc-SMIS) scheme with secure outsourcing computation for protecting and managing medical images in IoT environment. In the proposed scheme, the medical images are first encrypted using fully homomorphic encryption (FHE) and then outsourced to a cloud server for generating a set of image shares. Subsequently, these shares are stored separately in different local servers of medical institutions. Furthermore, the recovery process is also outsourced to the cloud server when doctors need to observe the patients’ medical images. Compared with the existing SMIS schemes, the proposed Enc-SMIS scheme alleviates the computing and communication burden on local servers significantly with secure outsourcing computation in the semi-honest model, and thus supports the storage and management of medical images well in IoT environment.


  • High-Capacity Steganography Using Object Addition-Based Cover Enhancement for Secure Communication in Networks

    by Ruohan Meng, Qi Cui, Zhili Zhou, Zhetao Li, Q. M. Jonathan Wu, Xingming Sun

    Steganography is an essential way to ensure secure communication in networks. Most steganographic algorithms imperceptibly embed secret information into an existing cover image. However, they generally cannot find a good trade-off between embedding capacity and security, as the existing covers available for users are usually far from optimal for embedding. To address this issue, instead of directly using the existing cover images, we propose a cover enhancement scheme for high-capacity image steganography, in which textured objects are generated and adaptively pasted to an existing cover based on the estimated embedding probability maps. Specifically, by estimating the embedding probability map of the cover image, we locate the high-embedding-cost region (HECR), which is inappropriate for embedding. Then, a textured object is generated by the conditional generative adversarial networks with the input of an affinely transformed object mask, and then is pasted to the located HECR for steganography. Since the image regions inappropriate for embedding are replaced by the textured object regions, the proposed scheme can provide a much higher embedding capacity for the state-of-the-art steganographic approaches. Extensive experiments demonstrate that the proposed scheme provides high embedding capacity, i.e. , about 2.5 times higher than the state-of-the-art steganographic methods, and comparable anti-detectability to those methods.