Jishen Zeng
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
Research Subjects
Jishen Zeng has a total of 32 co-authors. Here are the top co-authors by number of shared publications:
- Shunquan Tan
Shenzhen MSU-BIT University, Shenzhen, China
4 shared publications - Jianquan Yang
School of Cyber Science and Technology Shenzhen Campus of Sun Yat-sen University
3 shared publications - Jiwu Huang
Shenzhen MSU-BIT University, Shenzhen, China
3 shared publications Baoying Chen
Alibaba Group
2 shared publications- Bin Li
Shenzhen University, Shenzhen, Guangdong, China
2 shared publications - Dongliang Luo
Huazhong University of Science and Technology
2 shared publications
Latest Publications
- Open Access
Unveiling the Attribute Misbinding Threat in Identity-Preserving Models
Identity-preserving models have led to notable progress in generating personalized content. Unfortunately, such models also exacerbate risks when misused, for instance, by generating threatening content targeting specific individuals. This paper introduces the \textbf{Attribute Misbinding Attack}, a novel method that poses a threat to identity-preserving models by inducing them to produce Not-Safe-For-Work (NSFW) content. The attack's core idea involves crafting benign-looking textual prompts to circumvent text-filter safeguards and leverage a key model vulnerability: flawed attribute binding that stems from its internal attention bias. This results in misattributing harmful descriptions to a target identity and generating NSFW outputs. To facilitate the study of this attack, we present the \textbf{Misbinding Prompt} evaluation set, which examines the content generation risks of current state-of-the-art identity-preserving models across four risk dimensions: pornography, violence, discrimination, and illegality. Additionally, we introduce the \textbf{Attribute Binding Safety Score (ABSS)}, a metric for concurrently assessing both content fidelity and safety compliance. Experimental results show that our Misbinding Prompt evaluation set achieves a \textbf{5.28}\% higher success rate in bypassing five leading text filters (including GPT-4o) compared to existing main-stream evaluation sets, while also demonstrating the highest proportion of NSFW content generation. The proposed ABSS metric enables a more comprehensive evaluation of identity-preserving models by concurrently assessing both content fidelity and safety compliance.
Open AccessUnveiling the Attribute Misbinding Threat in Identity-Preserving Models
Identity-preserving models have led to notable progress in generating personalized content. Unfortunately, such models also exacerbate risks when misused, for instance, by generating threatening content targeting specific individuals. This paper introduces the \textbf{Attribute Misbinding Attack}, a novel method that poses a threat to identity-preserving models by inducing them to produce Not-Safe-For-Work (NSFW) content. The attack's core idea involves crafting benign-looking textual prompts to circumvent text-filter safeguards and leverage a key model vulnerability: flawed attribute binding that stems from its internal attention bias. This results in misattributing harmful descriptions to a target identity and generating NSFW outputs. To facilitate the study of this attack, we present the \textbf{Misbinding Prompt} evaluation set, which examines the content generation risks of current state-of-the-art identity-preserving models across four risk dimensions: pornography, violence, discrimination, and illegality. Additionally, we introduce the \textbf{Attribute Binding Safety Score (ABSS)}, a metric for concurrently assessing both content fidelity and safety compliance. Experimental results show that our Misbinding Prompt evaluation set achieves a \textbf{5.28}\% higher success rate in bypassing five leading text filters (including GPT-4o) compared to existing main-stream evaluation sets, while also demonstrating the highest proportion of NSFW content generation. The proposed ABSS metric enables a more comprehensive evaluation of identity-preserving models by concurrently assessing both content fidelity and safety compliance.
- Open Access
MSPT: A Lightweight Face Image Quality Assessment Method with Multi-stage Progressive Training
Accurately assessing the perceptual quality of face images is crucial, especially with the rapid progress in face restoration and generation. Traditional quality assessment methods often struggle with the unique characteristics of face images, limiting their generalizability. While learning-based approaches demonstrate superior performance due to their strong fitting capabilities, their high complexity typically incurs significant computational and storage costs, hindering practical deployment. To address this, we propose a lightweight face quality assessment network with Multi-Stage Progressive Training (MSPT). Our network employs a three-stage progressive training strategy that gradually introduces more diverse data samples and increases input image resolution. This novel approach enables lightweight networks to achieve high performance by effectively learning complex quality features while significantly mitigating catastrophic forgetting. Our MSPT achieved the second highest score on the VQualA 2025 face image quality assessment benchmark dataset, demonstrating that MSPT achieves comparable or better performance than state-of-the-art methods while maintaining efficient inference.
Open AccessMSPT: A Lightweight Face Image Quality Assessment Method with Multi-stage Progressive Training
Accurately assessing the perceptual quality of face images is crucial, especially with the rapid progress in face restoration and generation. Traditional quality assessment methods often struggle with the unique characteristics of face images, limiting their generalizability. While learning-based approaches demonstrate superior performance due to their strong fitting capabilities, their high complexity typically incurs significant computational and storage costs, hindering practical deployment. To address this, we propose a lightweight face quality assessment network with Multi-Stage Progressive Training (MSPT). Our network employs a three-stage progressive training strategy that gradually introduces more diverse data samples and increases input image resolution. This novel approach enables lightweight networks to achieve high performance by effectively learning complex quality features while significantly mitigating catastrophic forgetting. Our MSPT achieved the second highest score on the VQualA 2025 face image quality assessment benchmark dataset, demonstrating that MSPT achieves comparable or better performance than state-of-the-art methods while maintaining efficient inference.
- Open Access
- 5 May 2025
- Published by MDPI AG in Electronics
- Vol. 14 (9) , 1877
Multi-Scale Edge-Guided Image Forgery Detection via Improved Self-Supervision and Self-Adversarial Training
Image forgery detection, as an essential technique for analyzing image credibility, has experienced significant advancements recently. However, the forgery detection performance remains unsatisfactory in terms of meeting practical requirements. This is partly attributed to the limited availability of pixel-level annotated forgery samples and insufficient utilization of forgery traces. We try to mitigate these issues through three aspects: training data, network design, and training strategy. In the aspect of training data, we introduce iterative self-supervision which helps generate a large collection of pixel-level labeled single or composite forgery samples through one or more rounds of random copy-move, splicing, and inpainting, addressing the insufficient availability of forgery samples. In the aspect of network design, recognizing that characteristic anomalies are generally apparent at the boundary between true and fake regions, often aligning with image edges, we propose a new edge-guided learning module to effectively capture forgery traces at image edges. In the aspect of training strategy, we introduce progressive self-adversarial training, dynamically generating adversarial samples by gradually increasing the frequency and intensity of adversarial actions during training. This increases the detection difficulty, driving the detector to identify forgery traces from harder samples while maintaining a low computational cost. Comprehensive experiments have shown that the proposed method surpasses the leading competing methods, improving image-level forgery identification by 6.6% (from 73.8% to 80.4% on average F1 score) and pixel-level forgery localization by 15.2% (from 59.1% to 74.3% in average F1 score).
- 5 May 2025
- Published by MDPI AG in Electronics
- Vol. 14 (9) , 1877
Open AccessMulti-Scale Edge-Guided Image Forgery Detection via Improved Self-Supervision and Self-Adversarial Training
Image forgery detection, as an essential technique for analyzing image credibility, has experienced significant advancements recently. However, the forgery detection performance remains unsatisfactory in terms of meeting practical requirements. This is partly attributed to the limited availability of pixel-level annotated forgery samples and insufficient utilization of forgery traces. We try to mitigate these issues through three aspects: training data, network design, and training strategy. In the aspect of training data, we introduce iterative self-supervision which helps generate a large collection of pixel-level labeled single or composite forgery samples through one or more rounds of random copy-move, splicing, and inpainting, addressing the insufficient availability of forgery samples. In the aspect of network design, recognizing that characteristic anomalies are generally apparent at the boundary between true and fake regions, often aligning with image edges, we propose a new edge-guided learning module to effectively capture forgery traces at image edges. In the aspect of training strategy, we introduce progressive self-adversarial training, dynamically generating adversarial samples by gradually increasing the frequency and intensity of adversarial actions during training. This increases the detection difficulty, driving the detector to identify forgery traces from harder samples while maintaining a low computational cost. Comprehensive experiments have shown that the proposed method surpasses the leading competing methods, improving image-level forgery identification by 6.6% (from 73.8% to 80.4% on average F1 score) and pixel-level forgery localization by 15.2% (from 59.1% to 74.3% in average F1 score).
- 27 July 2024
- Published by Elsevier in Pattern Recognition
- Vol. 157
Toward real text manipulation detection: New dataset and new solution
- 27 July 2024
- Published by Elsevier in Pattern Recognition
- Vol. 157
Toward real text manipulation detection: New dataset and new solution
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1556-1564
On Physically Occluded Fake Identity Document Detection
Many online applications require the users to upload their identity documents for authentication. The fake identity document is one of the main threats which compromises the security and reliability of such online applications. Existing techniques focus on the detection of digitally forged identity documents, which neglect the impact of physical forgeries. In this paper, we look into the problem of detecting physically occluded fake identity documents, which can be easily generated without any image processing knowledge. We observe that the physical occlusions inevitably produce occluded boundaries on the document. To take the advantage, we propose an Occluded Boundary Representation Learning (OBRL) module to progressively learn the occluded boundary features. These are then fed into an Occluded Boundary Message Passing (OBMP) module to effectively diffuse the physical occlusion traces to enhance the backbone features for robust detection. We newly construct a Physically Occluded Fake ID Card image dataset (POID) for evaluation. Various experiments are conducted on the POID, where our scheme is able to achieve 99.6% of accuracy in detecting physically occluded fake ID card images with a mAP of over 85% to localize the occlusion regions.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 1556-1564
Open AccessOn Physically Occluded Fake Identity Document Detection
Many online applications require the users to upload their identity documents for authentication. The fake identity document is one of the main threats which compromises the security and reliability of such online applications. Existing techniques focus on the detection of digitally forged identity documents, which neglect the impact of physical forgeries. In this paper, we look into the problem of detecting physically occluded fake identity documents, which can be easily generated without any image processing knowledge. We observe that the physical occlusions inevitably produce occluded boundaries on the document. To take the advantage, we propose an Occluded Boundary Representation Learning (OBRL) module to progressively learn the occluded boundary features. These are then fed into an Occluded Boundary Message Passing (OBMP) module to effectively diffuse the physical occlusion traces to enhance the backbone features for robust detection. We newly construct a Physically Occluded Fake ID Card image dataset (POID) for evaluation. Various experiments are conducted on the POID, where our scheme is able to achieve 99.6% of accuracy in detecting physically occluded fake ID card images with a mAP of over 85% to localize the occlusion regions.
- 19 August 2023
- Published by Springer Nature
- p. 587-600
ICDAR 2023 Competition on Detecting Tampered Text in Images
- 19 August 2023
- Published by Springer Nature
- p. 587-600
ICDAR 2023 Competition on Detecting Tampered Text in Images
- Open Access
- 30 November 2020
- Published by Korean Society for Internet Information (KSII) in KSII Transactions on Internet and Information Systems
- Vol. 14 (11) , 4557-4572
Ensemble Deep Learning Features for Real-World Image Steganalysis
KSII Transactions on Internet and Information Systems TIIS Paper Details
- 30 November 2020
- Published by Korean Society for Internet Information (KSII) in KSII Transactions on Internet and Information Systems
- Vol. 14 (11) , 4557-4572
Open AccessEnsemble Deep Learning Features for Real-World Image Steganalysis
KSII Transactions on Internet and Information Systems TIIS Paper Details
- 11 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 14 (10) , 2735-2748
WISERNet: Wider Separate-Then-Reunion Network for Steganalysis of Color Images
Until recently, deep steganalyzers in spatial domain have been all designed for gray-scale images. In this paper, we propose WISERNet (the wider separate-then-reunion network) for steganalysis of color images. We provide theoretical rationale to claim that the summation in normal convolution is one sort of “linear collusion attack” which reserves strong correlated patterns while impairs uncorrelated noises. Therefore in the bottom convolutional layer which aims at suppressing correlated image contents, we adopt separate channel-wise convolution without summation instead. Conversely, in the upper convolutional layers we believe that the summation in normal convolution is beneficial. Therefore we adopt united normal convolution in those layers and make them remarkably wider to reinforce the effect of “linear collusion attack”. As a result, our proposed wide-and-shallow, separate-then-reunion network structure is specifically suitable for color image steganalysis. We have conducted extensive experiments on color image datasets generated from BOSSBase raw images and another large-scale dataset which contains 100,000 raw images, with different demosaicking algorithms and downsampling algorithms. The experimental results show that our proposed network outperforms other state-of-the-art color image steganalytic models either hand-crafted or learned using deep networks in the literature by a clear margin. Specifically, it is noted that the detection performance gain is achieved with less than half the complexity compared to the most advanced deeplearning steganalyzer as far as we know, which is scarce in the literature.
- 11 March 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 14 (10) , 2735-2748
WISERNet: Wider Separate-Then-Reunion Network for Steganalysis of Color Images
Until recently, deep steganalyzers in spatial domain have been all designed for gray-scale images. In this paper, we propose WISERNet (the wider separate-then-reunion network) for steganalysis of color images. We provide theoretical rationale to claim that the summation in normal convolution is one sort of “linear collusion attack” which reserves strong correlated patterns while impairs uncorrelated noises. Therefore in the bottom convolutional layer which aims at suppressing correlated image contents, we adopt separate channel-wise convolution without summation instead. Conversely, in the upper convolutional layers we believe that the summation in normal convolution is beneficial. Therefore we adopt united normal convolution in those layers and make them remarkably wider to reinforce the effect of “linear collusion attack”. As a result, our proposed wide-and-shallow, separate-then-reunion network structure is specifically suitable for color image steganalysis. We have conducted extensive experiments on color image datasets generated from BOSSBase raw images and another large-scale dataset which contains 100,000 raw images, with different demosaicking algorithms and downsampling algorithms. The experimental results show that our proposed network outperforms other state-of-the-art color image steganalytic models either hand-crafted or learned using deep networks in the literature by a clear margin. Specifically, it is noted that the detection performance gain is achieved with less than half the complexity compared to the most advanced deeplearning steganalyzer as far as we know, which is scarce in the literature.
- 4 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 13 (5) , 1200-1214
Large-Scale JPEG Image Steganalysis Using Hybrid Deep-Learning Framework
Adoption of deep learning in image steganalysis is still in its initial stage. In this paper, we propose a generic hybrid deep-learning framework for JPEG steganalysis incorporating the domain knowledge behind rich steganalytic models. Our proposed framework involves two main stages. The first stage is hand-crafted, corresponding to the convolution phase and the quantization and truncation phase of the rich models. The second stage is a compound deep-neural network containing multiple deep subnets, in which the model parameters are learned in the training procedure. We provided experimental evidence and theoretical reflections to argue that the introduction of threshold quantizers, though disabling the gradient-descent-based learning of the bottom convolution phase, is indeed cost-effective. We have conducted extensive experiments on a large-scale data set extracted from ImageNet. The primary data set used in our experiments contains 500 000 cover images, while our largest data set contains five million cover images. Our experiments show that the integration of quantization and truncation into deep-learning steganalyzers do boost the detection performance by a clear margin. Furthermore, we demonstrate that our framework is insensitive to JPEG blocking artifact alterations, and the learned model can be easily transferred to a different attacking target and even a different data set. These properties are of critical importance in practical applications.
- 4 December 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Forensics and Security
- Vol. 13 (5) , 1200-1214
Large-Scale JPEG Image Steganalysis Using Hybrid Deep-Learning Framework
Adoption of deep learning in image steganalysis is still in its initial stage. In this paper, we propose a generic hybrid deep-learning framework for JPEG steganalysis incorporating the domain knowledge behind rich steganalytic models. Our proposed framework involves two main stages. The first stage is hand-crafted, corresponding to the convolution phase and the quantization and truncation phase of the rich models. The second stage is a compound deep-neural network containing multiple deep subnets, in which the model parameters are learned in the training procedure. We provided experimental evidence and theoretical reflections to argue that the introduction of threshold quantizers, though disabling the gradient-descent-based learning of the bottom convolution phase, is indeed cost-effective. We have conducted extensive experiments on a large-scale data set extracted from ImageNet. The primary data set used in our experiments contains 500 000 cover images, while our largest data set contains five million cover images. Our experiments show that the integration of quantization and truncation into deep-learning steganalyzers do boost the detection performance by a clear margin. Furthermore, we demonstrate that our framework is insensitive to JPEG blocking artifact alterations, and the learned model can be easily transferred to a different attacking target and even a different data set. These properties are of critical importance in practical applications.
- 29 January 2017
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 29 (7) , 44-49
Pre-training via fitting deep neural network to rich-model features extraction procedure and its effect on deep learning for steganalysis
Recent studies have shown that the steganalytic approaches based on deep learning frameworks cannot surpass their rich?model features based companions in peiformance. According to our analysis, one of the main causes of the unsatisfactory performance of deep learning frameworks is that training procedure tends to get stuck at local plateaus or even diverge when starting from a non-ideal initial state. In this paper we will try to investi?gate how to fit deep neural network to a rich-model features set. We regard it as a pre-training procedure and study its 4fect on deep learning for steganalysis. The state-of-the-art JPEG steganalytic features set DCTR is selected as the target and its features extraction procedure is divided into multiple sub-models. A deep learning framework with similar sub-networks is proposed. In the pre-training procedure we train theframeworkfrom bottom to up, fitting the output of each sub-network to the actual output of the corresponding sub-module of DCTR. The motivation behind the scenario is that we reinforce the proposed framework learn to fit the nonlinear mapping implicit in DCTR and expect when it is trainedfrom an initial state which represents an approximate so?lution of DCTR, we can get better peiformance compared to what DCTR has achieved.
- 29 January 2017
- Published by Society for Imaging Science & Technology in Electronic Imaging
- Vol. 29 (7) , 44-49
Pre-training via fitting deep neural network to rich-model features extraction procedure and its effect on deep learning for steganalysis
Recent studies have shown that the steganalytic approaches based on deep learning frameworks cannot surpass their rich?model features based companions in peiformance. According to our analysis, one of the main causes of the unsatisfactory performance of deep learning frameworks is that training procedure tends to get stuck at local plateaus or even diverge when starting from a non-ideal initial state. In this paper we will try to investi?gate how to fit deep neural network to a rich-model features set. We regard it as a pre-training procedure and study its 4fect on deep learning for steganalysis. The state-of-the-art JPEG steganalytic features set DCTR is selected as the target and its features extraction procedure is divided into multiple sub-models. A deep learning framework with similar sub-networks is proposed. In the pre-training procedure we train theframeworkfrom bottom to up, fitting the output of each sub-network to the actual output of the corresponding sub-module of DCTR. The motivation behind the scenario is that we reinforce the proposed framework learn to fit the nonlinear mapping implicit in DCTR and expect when it is trainedfrom an initial state which represents an approximate so?lution of DCTR, we can get better peiformance compared to what DCTR has achieved.