Weixuan Tang

Ao-Xiang Zhang, Yu Ran, Weixuan Tang et al.

No-Reference Video Quality Assessment (NR-VQA) plays an essential role in improving the viewing experience of end-users. Driven by deep learning, recent NR-VQA models based on Convolutional Neural Networks (CNNs) and Transformers have achieved outstanding performance. To build a reliable and practical assessment system, it is of great necessity to evaluate their robustness. However, such issue has received little attention in the academic community. In this paper, we make the first attempt to evaluate the robustness of NR-VQA models against adversarial attacks, and propose a patch-based random search method for black-box attack. Specifically, considering both the attack effect on quality score and the visual quality of adversarial video, the attack problem is formulated as misleading the estimated quality score under the constraint of just-noticeable difference (JND). Built upon such formulation, a novel loss function called Score-Reversed Boundary Loss is designed to push the adversarial video's estimated quality score far away from its ground-truth score towards a specific boundary, and the JND constraint is modeled as a strict $L_2$ and $L_\infty$ norm restriction. By this means, both white-box and black-box attacks can be launched in an effective and imperceptible manner. The source code is available at https://github.com/GZHU-DVL/AttackVQA.

Ao-Xiang Zhang, Yuan-Gen Wang, Weixuan Tang et al.

Video quality is a primary concern for video service providers. In recent years, the techniques of video quality assessment (VQA) based on deep convolutional neural networks (CNNs) have been developed rapidly. Although existing works attempt to introduce the knowledge of the human visual system (HVS) into VQA, there still exhibit limitations that prevent the full exploitation of HVS, including an incomplete model by few characteristics and insufficient connections among these characteristics. To overcome these limitations, this paper revisits HVS with five representative characteristics, and further reorganizes their connections. Based on the revisited HVS, a no-reference VQA framework called HVS-5M (NRVQA framework with five modules simulating HVS with five characteristics) is proposed. It works in a domain-fusion design paradigm with advanced network structures. On the side of the spatial domain, the visual saliency module applies SAMNet to obtain a saliency map. And then, the content-dependency and the edge masking modules respectively utilize ConvNeXt to extract the spatial features, which have been attentively weighted by the saliency map for the purpose of highlighting those regions that human beings may be interested in. On the other side of the temporal domain, to supplement the static spatial features, the motion perception module utilizes SlowFast to obtain the dynamic temporal features. Besides, the temporal hysteresis module applies TempHyst to simulate the memory mechanism of human beings, and comprehensively evaluates the quality score according to the fusion features from the spatial and temporal domains. Extensive experiments show that our HVS-5M outperforms the state-of-the-art VQA methods. Ablation studies are further conducted to verify the effectiveness of each module towards the proposed framework.

Fengchuang Xing, Yuan-Gen Wang, Weixuan Tang et al.

Self-attention based Transformer has achieved great success in many computer vision tasks. However, its application to video quality assessment (VQA) has not been satisfactory so far. Evaluating the quality of in-the-wild videos is challenging due to the unknown of pristine reference and shooting distortion. This paper presents a co-trained Space-Time Attention network for the VQA problem, termed StarVQA+. Specifically, we first build StarVQA+ by alternately concatenating the divided space-time attention. Then, to facilitate the training of StarVQA+, we design a vectorized regression loss by encoding the mean opinion score (MOS) to the probability vector and embedding a special token as the learnable variable of MOS, leading to better fitting of human's rating process. Finally, to solve the data hungry problem with Transformer, we propose to co-train the spatial and temporal attention weights using both images and videos. Various experiments are conducted on the de-facto in-the-wild video datasets, including LIVE-Qualcomm, LIVE-VQC, KoNViD-1k, YouTube-UGC, LSVQ, LSVQ-1080p, and DVL2021. Experimental results demonstrate the superiority of the proposed StarVQA+ over the state-of-the-art.

Wei Wu, Xi Guo, Weixuan Tang et al.

Recent advancements in generative models have provided promising solutions for synthesizing realistic driving videos, which are crucial for training autonomous driving perception models. However, existing approaches often struggle with multi-view video generation due to the challenges of integrating 3D information while maintaining spatial-temporal consistency and effectively learning from a unified model. We propose DriveScape, an end-to-end framework for multi-view, 3D condition-guided video generation, capable of producing 1024 x 576 high-resolution videos at 10Hz. Unlike other methods limited to 2Hz due to the 3D box annotation frame rate, DriveScape overcomes this with its ability to operate under sparse conditions. Our Bi-Directional Modulated Transformer (BiMot) ensures precise alignment of 3D structural information, maintaining spatial-temporal consistency. DriveScape excels in video generation performance, achieving state-of-the-art results on the nuScenes dataset with an FID score of 8.34 and an FVD score of 76.39. Our project homepage: https://metadrivescape.github.io/papers_project/drivescapev1/index.html

Chenjing Ding, Chiyu Wang, Boshi Liu et al.

Vector quantization (VQ) is a method for deterministically learning features through discrete codebook representations. Recent works have utilized visual tokenizers to discretize visual regions for self-supervised representation learning. However, a notable limitation of these tokenizers is lack of semantics, as they are derived solely from the pretext task of reconstructing raw image pixels in an auto-encoder paradigm. Additionally, issues like imbalanced codebook distribution and codebook collapse can adversely impact performance due to inefficient codebook utilization. To address these challenges, We introduce SGC-VQGAN through Semantic Online Clustering method to enhance token semantics through Consistent Semantic Learning. Utilizing inference results from segmentation model , our approach constructs a temporospatially consistent semantic codebook, addressing issues of codebook collapse and imbalanced token semantics. Our proposed Pyramid Feature Learning pipeline integrates multi-level features to capture both image details and semantics simultaneously. As a result, SGC-VQGAN achieves SOTA performance in both reconstruction quality and various downstream tasks. Its simplicity, requiring no additional parameter learning, enables its direct application in downstream tasks, presenting significant potential.

Yu Ran, Ao-Xiang Zhang, Mingjie Li et al.

The goal of No-Reference Image Quality Assessment (NR-IQA) is to predict the perceptual quality of an image in line with its subjective evaluation. To put the NR-IQA models into practice, it is essential to study their potential loopholes for model refinement. This paper makes the first attempt to explore the black-box adversarial attacks on NR-IQA models. Specifically, we first formulate the attack problem as maximizing the deviation between the estimated quality scores of original and perturbed images, while restricting the perturbed image distortions for visual quality preservation. Under such formulation, we then design a Bi-directional loss function to mislead the estimated quality scores of adversarial examples towards an opposite direction with maximum deviation. On this basis, we finally develop an efficient and effective black-box attack method against NR-IQA models. Extensive experiments reveal that all the evaluated NR-IQA models are vulnerable to the proposed attack method. And the generated perturbations are not transferable, enabling them to serve the investigation of specialities of disparate IQA models.

Xi Guo, Chenjing Ding, Haoxuan Dou et al.

Autonomous driving systems struggle with complex scenarios due to limited access to diverse, extensive, and out-of-distribution driving data which are critical for safe navigation. World models offer a promising solution to this challenge; however, current driving world models are constrained by short time windows and limited scenario diversity. To bridge this gap, we introduce InfinityDrive, the first driving world model with exceptional generalization capabilities, delivering state-of-the-art performance in high fidelity, consistency, and diversity with minute-scale video generation. InfinityDrive introduces an efficient spatio-temporal co-modeling module paired with an extended temporal training strategy, enabling high-resolution (576$\times$1024) video generation with consistent spatial and temporal coherence. By incorporating memory injection and retention mechanisms alongside an adaptive memory curve loss to minimize cumulative errors, achieving consistent video generation lasting over 1500 frames (more than 2 minutes). Comprehensive experiments in multiple datasets validate InfinityDrive's ability to generate complex and varied scenarios, highlighting its potential as a next-generation driving world model built for the evolving demands of autonomous driving. Our project homepage: https://metadrivescape.github.io/papers_project/InfinityDrive/page.html

Junxue Yang, Xin Liao, Weixuan Tang et al.

Image hiding fully explores the hidden potential of deep learning-based models, aiming to conceal image-level messages within cover images and reveal them from stego images to achieve covert communication. Existing hiding schemes are easily detected by the naked eyes or steganalyzers due to the cover type confined to the spatial domain, single-range feature extraction and attacks, and insufficient loss constraints. To address these issues, we propose a multi-range representations-driven adversarial stego generation framework called MRAG for JPEG image hiding. This design stems from the fact that steganalyzers typically combine local-range and global-range information to better capture hidden traces. Specifically, MRAG integrates the local-range characteristic of the convolution and the global-range modeling of the transformer. Meanwhile, a features angle-norm disentanglement loss is designed to launch multi-range representations-driven feature-level adversarial attacks. It computes the adversarial loss between covers and stegos based on the surrogate steganalyzer's classified features, i.e., the features before the last fully connected layer. Under the dual constraints of features angle and norm, MRAG can delicately encode the concatenation of cover and secret into subtle adversarial perturbations from local and global ranges relevant to steganalysis. Therefore, the resulting stego can achieve visual and steganalysis imperceptibility. Moreover, coarse-grained and fine-grained frequency decomposition operations are devised to transform the input, introducing multi-grained information. Extensive experiments demonstrate that MRAG can achieve state-of-the-art performance.

Weixuan Tang, Danping Zou

We address the problem of estimating the poses of multiple instances of the source point cloud within a target point cloud. Existing solutions require sampling a lot of hypotheses to detect possible instances and reject the outliers, whose robustness and efficiency degrade notably when the number of instances and outliers increase. We propose to directly group the set of noisy correspondences into different clusters based on a distance invariance matrix. The instances and outliers are automatically identified through clustering. Our method is robust and fast. We evaluated our method on both synthetic and real-world datasets. The results show that our approach can correctly register up to 20 instances with an F1 score of 90.46% in the presence of 70% outliers, which performs significantly better and at least 10x faster than existing methods

Weixuan Tang, Bin Li, Mauro Barni et al.

Although significant progress in automatic learning of steganographic cost has been achieved recently, existing methods designed for spatial images are not well applicable to JPEG images which are more common media in daily life. The difficulties of migration mostly lie in the unique and complicated JPEG characteristics caused by 8x8 DCT mode structure. To address the issue, in this paper we extend an existing automatic cost learning scheme to JPEG, where the proposed scheme called JEC-RL (JPEG Embedding Cost with Reinforcement Learning) is explicitly designed to tailor the JPEG DCT structure. It works with the embedding action sampling mechanism under reinforcement learning, where a policy network learns the optimal embedding policies via maximizing the rewards provided by an environment network. The policy network is constructed following a domain-transition design paradigm, where three modules including pixel-level texture complexity evaluation, DCT feature extraction, and mode-wise rearrangement, are proposed. These modules operate in serial, gradually extracting useful features from a decompressed JPEG image and converting them into embedding policies for DCT elements, while considering JPEG characteristics including inter-block and intra-block correlations simultaneously. The environment network is designed in a gradient-oriented way to provide stable reward values by using a wide architecture equipped with a fixed preprocessing layer with 8x8 DCT basis filters. Extensive experiments and ablation studies demonstrate that the proposed method can achieve good security performance for JPEG images against both advanced feature based and modern CNN based steganalyzers.

Xinghong Qin, Shunquan Tan, Bin Li et al.

Nowadays a steganography has to face challenges of both feature based staganalysis and convolutional neural network (CNN) based steganalysis. In this paper, we present a novel steganography scheme denoted as ITE-SYN (based on ITEratively adversarial perturbations onto a SYNchronized-directions sub-image), by which security data is embedded with synchronizing modification directions to enhance security and then iteratively increased perturbations are added onto a sub-image to reduce loss with cover class label of the target CNN classifier. Firstly an exist steganographic function is employed to compute initial costs. Then the cover image is decomposed into some non-overlapped sub-images. After each sub-image is embedded, costs will be adjusted following clustering modification directions profile. And then the next sub-image will be embedded with adjusted costs until all secret data has been embedded. If the target CNN classifier does not discriminate the stego image as a cover image, based on adjusted costs, we change costs with adversarial manners according to signs of gradients back-propagated from the CNN classifier. And then a sub-image is chosen to be re-embedded with changed costs. Adversarial intensity will be iteratively increased until the adversarial stego image can fool the target CNN classifier. Experiments demonstrate that the proposed method effectively enhances security to counter both conventional feature-based classifiers and CNN classifiers, even other non-target CNN classifiers.

Weixuan Tang, Bin Li, Shunquan Tan et al.

Historically, steganographic schemes were designed in a way to preserve image statistics or steganalytic features. Since most of the state-of-the-art steganalytic methods employ a machine learning (ML) based classifier, it is reasonable to consider countering steganalysis by trying to fool the ML classifiers. However, simply applying perturbations on stego images as adversarial examples may lead to the failure of data extraction and introduce unexpected artefacts detectable by other classifiers. In this paper, we present a steganographic scheme with a novel operation called adversarial embedding, which achieves the goal of hiding a stego message while at the same time fooling a convolutional neural network (CNN) based steganalyzer. The proposed method works under the conventional framework of distortion minimization. Adversarial embedding is achieved by adjusting the costs of image element modifications according to the gradients backpropagated from the CNN classifier targeted by the attack. Therefore, modification direction has a higher probability to be the same as the sign of the gradient. In this way, the so called adversarial stego images are generated. Experiments demonstrate that the proposed steganographic scheme is secure against the targeted adversary-unaware steganalyzer. In addition, it deteriorates the performance of other adversary-aware steganalyzers opening the way to a new class of modern steganographic schemes capable to overcome powerful CNN-based steganalysis.