Sanshuai Cui

Dr. Sanshuai Cui

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
2 March 2023 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
BHBingyuan Huang
Bingyuan Huang
SCSanshuai Cui
Sanshuai Cui
JHJiwu Huang
Jiwu Huang
XKXiangui Kang
Xiangui Kang
End-to-end technology is an active research topic in speech anti-spoofing. Although end-to-end methods have achieved remarkable success in the speech anti-spoofing, channel effects brought by telephony transmission and certain challenging forms of spoofing attacks still plague them. We observe that differences in the high-frequency components between bonafide and spoofed speech help detect some most troublesome attack forms and the differences also remain after the signals are affected by transmission and codecs. Based on this observation, we aim to utilize the high-frequency information of speech signals to develop better generalization ability to unknown attacks and stronger robustness against transmission and codecs. We propose a raw waveform processing module based on sinc convolution and multiple pre-emphasis to obtain discriminative shallow feature representations. Additionally, we propose an improved backbone to learn discriminative feature embeddings, and a feature classification loss to optimize intra-class and inter-class distances simultaneously. The above modules constitute the proposed Discriminative Frequency-information SincNet, namely DFSincNet. Our proposed algorithm demonstrates competitive performance on both ASVspoof 2019 and 2021 logical access (LA) scenarios.

17 June 2022 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
SCSanshuai Cui
Sanshuai Cui
BHBingyuan Huang
Bingyuan Huang
JHJiwu Huang
Jiwu Huang
XKXiangui Kang
Xiangui Kang
With the development of speech synthesis technology, the existing synthetic speech detection (SSD) methods cannot generalize well for unknown synthesis algorithms. And thus, this kind of speech forensics task meets a great challenge and has attracted great enthusiasm. We observe that the process of speech synthesis always includes the resampling and pooling/smoothing operations, which will change the speech’s local autoregressive (AR) and statistic distribution. In this paper, based on AR modeling and standard deviation statistics, we propose novel front-end speech features, i.e., ARS in short form, as the input of an SSD classifier. In addition, a new back-end classifier is constructed based on the dense convolution and short connection, and we name it scDenseNet. Experimental results on the ASVspoof2019 logical access (LA) dataset demonstrate that the ARS has a strong representation and sensitivity to spoofing attacks, and achieves promising performance on SSD. The proposed scDenseNet outperforms the previous version DenseNet on both EER and t-DCF scores, and achieves the best performance when compared with other state-of-the-art classifiers studied in this paper. Furthermore, based on the proposed scDenseNet, incorporating ARS with popular features such as the linear frequency cepstral coefficients (LFCC) significantly enhances the fusion performance and yields an EER score of 0.98%.

2 October 2024 in Mathematics
Verified
Journal Article
Open access
PDF file
XYXudong Ye
Xudong Ye
Qi ZhangQi Zhang
Qi Zhang
SCSanshuai Cui
Sanshuai Cui
Zuobin YingZuobin Ying
Zuobin Ying
JSJingzhang Sun
Jingzhang Sun
Xia DuXia Du
Xia Du
The field of object detection has witnessed significant advancements in recent years, thanks to the remarkable progress in artificial intelligence and deep learning. These breakthroughs have significantly enhanced the accuracy and efficiency of detecting and categorizing objects in digital images. Nonetheless, contemporary object detection technologies have certain limitations, such as their inability to counter white-box attacks, insufficient denoising, suboptimal reconstruction, and gradient confusion. To overcome these hurdles, this study proposes an innovative approach that uses conditional diffusion models to perturb adversarial examples. The process begins with the application of a random chessboard mask to the adversarial example, followed by the addition of a slight noise to fill the masked area during the forward process. The adversarial image is then restored to its original form through a reverse generative process that only considers the masked pixels, not the entire image. Next, we use the complement of the initial mask as the mask for the second stage to reconstruct the image once more. This two-stage masking process allows for the complete removal of global disturbances and aids in image reconstruction. In particular, we employ a conditional diffusion model based on a class-conditional U-Net architecture, with the source image further conditioned through concatenation. Our method outperforms the recently introduced HARP method by 5% and 6.5% in mAP on the COCO2017 and PASCAL VOC datasets, respectively, under non-APT PGD attacks. Comprehensive experimental results confirm that our method can effectively restore adversarial examples, demonstrating its practical utility.

23 April 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
SCSanshuai Cui
Sanshuai Cui
SCSanshuai Cui
Sanshuai Cui
Qi ZhangQi Zhang
Qi Zhang
BCBIWEI CHEN
BIWEI CHEN
HZHui Zeng
Hui Zeng
QZQi Zhong
Qi Zhong
In recent years, with the rapid advancement of deep learning technologies such as generative adversarial networks (GANs), deepfake technology has become increasingly sophisticated. As a result, the generated fake images are becoming more difficult to visually distinguish from real ones. Existing deepfake detection methods primarily rely on training models with specific datasets. However, these models often suffer from limited generalization when processing images of unknown origin or across domains, leading to a significant decrease in detection accuracy. To address this issue, this paper proposes a deepfake image-detection network based on feature aggregation and enhancement. The key innovation of the proposed method lies in the integration of two modules: the Feature Aggregation Module (FAM) and the Attention Enhancement Module (AEM). The FAM effectively aggregates both deep semantic information and shallow detail features through a multi-scale feature-fusion mechanism, overcoming the limitations of traditional methods that rely on a single-level feature. Meanwhile, the AEM enhances the network’s ability to capture subtle forgery traces by incorporating attention mechanisms and filtering techniques, significantly boosting the model’s efficiency in processing complex information. The experimental results demonstrate that the proposed method achieves significant improvements across all evaluation metrics. Specifically, on the StarGAN dataset, the model attained outstanding performance, with accuracy (Acc) and average precision (AP) both reaching 100%. In cross-dataset testing, the proposed method exhibited strong generalization ability, raising the overall average accuracy to 87.0% and average precision to 92.8%, representing improvements of 5.2% and 6.7%, respectively, compared to existing state-of-the-art methods. These results show that the proposed method can not only achieve optimal performance on data with the same distribution, but also demonstrate strong generalization ability in cross-domain detection tasks.


11 April 2025 in Proceedings of the AAAI Conference on Artificial Intelligence
Verified
Journal Article
Open access
PDF file
HZHui Zeng
Hui Zeng
SCSanshuai Cui
Sanshuai Cui
BCBIWEI CHEN
BIWEI CHEN
APAnjie Peng
Anjie Peng
Adversarial examples’ (AE) transferability refers to the phenomenon that AEs crafted with one surrogate model can also fool other models. Notwithstanding remarkable progress in untargeted transferability, its targeted counterpart remains challenging. This paper proposes an everywhere scheme to boost targeted transferability. Our idea is to attack a victim image both globally and locally. We aim to optimize ‘an army of targets’ in every local image region instead of the previous works that optimize a high-confidence target in the image. Specifically, we split a victim image into non-overlap blocks and jointly mount a targeted attack on each block. Such a strategy mitigates transfer failures caused by attention inconsistency between surrogate and victim models and thus results in stronger transferability. Our approach is method-agnostic, which means it can be easily combined with existing transferable attacks for even higher transferability. Extensive experiments on ImageNet demonstrate that the proposed approach universally improves the state-of-the-art targeted attacks by a clear margin, e.g., the transferability of the widely adopted Logit attack can be improved by 28.8%-300%. We also evaluate the crafted AEs on a real-world platform: Google Cloud Vision. Results further support the superiority of the proposed method.