Sheng Zhong

Qiwen Zhu, Yanjie Wang, Shilv Cai et al.

Training Single-Image Super-Resolution (SISR) models using pixel-based regression losses can achieve high distortion metrics scores (e.g., PSNR and SSIM), but often results in blurry images due to insufficient recovery of high-frequency details. Conversely, using GAN or perceptual losses can produce sharp images with high perceptual metric scores (e.g., LPIPS), but may introduce artifacts and incorrect textures. Balancing these two types of losses can help achieve a trade-off between distortion and perception, but the challenge lies in tuning the loss function weights. To address this issue, we propose a novel method that incorporates Multi-Objective Optimization (MOO) into the training process of SISR models to balance perceptual quality and distortion. We conceptualize the relationship between loss weights and image quality assessment (IQA) metrics as black-box objective functions to be optimized within our Multi-Objective Bayesian Optimization Super-Resolution (MOBOSR) framework. This approach automates the hyperparameter tuning process, reduces overall computational cost, and enables the use of numerous loss functions simultaneously. Extensive experiments demonstrate that MOBOSR outperforms state-of-the-art methods in terms of both perceptual quality and distortion, significantly advancing the perception-distortion Pareto frontier. Our work points towards a new direction for future research on balancing perceptual quality and fidelity in nearly all image restoration tasks. The source code and pretrained models are available at: https://github.com/ZhuKeven/MOBOSR.

Shilv Cai, Zhijun Zhang, Liqun Chen et al.

Learning-based methods have effectively promoted the community of image compression. Meanwhile, variational autoencoder (VAE) based variable-rate approaches have recently gained much attention to avoid the usage of a set of different networks for various compression rates. Despite the remarkable performance that has been achieved, these approaches would be readily corrupted once multiple compression/decompression operations are executed, resulting in the fact that image quality would be tremendously dropped and strong artifacts would appear. Thus, we try to tackle the issue of high-fidelity fine variable-rate image compression and propose the Invertible Activation Transformation (IAT) module. We implement the IAT in a mathematical invertible manner on a single rate Invertible Neural Network (INN) based model and the quality level (QLevel) would be fed into the IAT to generate scaling and bias tensors. IAT and QLevel together give the image compression model the ability of fine variable-rate control while better maintaining the image fidelity. Extensive experiments demonstrate that the single rate image compression model equipped with our IAT module has the ability to achieve variable-rate control without any compromise. And our IAT-embedded model obtains comparable rate-distortion performance with recent learning-based image compression methods. Furthermore, our method outperforms the state-of-the-art variable-rate image compression method by a large margin, especially after multiple re-encodings.

Songrui Wang, Yubo Zhu, Wei Tong et al.

Text-to-image synthesis has become highly popular for generating realistic and stylized images, often requiring fine-tuning generative models with domain-specific datasets for specialized tasks. However, these valuable datasets face risks of unauthorized usage and unapproved sharing, compromising the rights of the owners. In this paper, we address the issue of dataset abuse during the fine-tuning of Stable Diffusion models for text-to-image synthesis. We present a dataset watermarking framework designed to detect unauthorized usage and trace data leaks. The framework employs two key strategies across multiple watermarking schemes and is effective for large-scale dataset authorization. Extensive experiments demonstrate the framework's effectiveness, minimal impact on the dataset (only 2% of the data required to be modified for high detection accuracy), and ability to trace data leaks. Our results also highlight the robustness and transferability of the framework, proving its practical applicability in detecting dataset abuse.

Junkai Niu, Sheng Zhong, Xiuyuan Lu et al.

Event-based visual odometry is a specific branch of visual Simultaneous Localization and Mapping (SLAM) techniques, which aims at solving tracking and mapping subproblems (typically in parallel), by exploiting the special working principles of neuromorphic (i.e., event-based) cameras. Due to the motion-dependent nature of event data, explicit data association (i.e., feature matching) under large-baseline view-point changes is difficult to establish, making direct methods a more rational choice. However, state-of-the-art direct methods are limited by the high computational complexity of the mapping sub-problem and the degeneracy of camera pose tracking in certain degrees of freedom (DoF) in rotation. In this paper, we tackle these issues by building an event-based stereo visual-inertial odometry system on top of a direct pipeline. Specifically, to speed up the mapping operation, we propose an efficient strategy for sampling contour points according to the local dynamics of events. The mapping performance is also improved in terms of structure completeness and local smoothness by merging the temporal stereo and static stereo results. To circumvent the degeneracy of camera pose tracking in recovering the pitch and yaw components of general 6-DoF motion, we introduce IMU measurements as motion priors via pre-integration. To this end, a compact back-end is proposed for continuously updating the IMU bias and predicting the linear velocity, enabling an accurate motion prediction for camera pose tracking. The resulting system scales well with modern high-resolution event cameras and leads to better global positioning accuracy in large-scale outdoor environments. Extensive evaluations on five publicly available datasets featuring different resolutions and scenarios justify the superior performance of the proposed system against five state-of-the-art methods.

Shilv Cai, Xiaoguo Liang, Shuning Cao et al.

Image compression and denoising represent fundamental challenges in image processing with many real-world applications. To address practical demands, current solutions can be categorized into two main strategies: 1) sequential method; and 2) joint method. However, sequential methods have the disadvantage of error accumulation as there is information loss between multiple individual models. Recently, the academic community began to make some attempts to tackle this problem through end-to-end joint methods. Most of them ignore that different regions of noisy images have different characteristics. To solve these problems, in this paper, our proposed signal-to-noise ratio~(SNR) aware joint solution exploits local and non-local features for image compression and denoising simultaneously. We design an end-to-end trainable network, which includes the main encoder branch, the guidance branch, and the signal-to-noise ratio~(SNR) aware branch. We conducted extensive experiments on both synthetic and real-world datasets, demonstrating that our joint solution outperforms existing state-of-the-art methods.

Tianyu Zhang, Zihang Xi, Jingyu Hua et al.

In the realm of black-box jailbreak attacks on large language models (LLMs), the feasibility of constructing a narrow safety proxy, a lightweight model designed to predict the attack success rate (ASR) of adversarial prompts, remains underexplored. This work investigates the distillability of an LLM's core security logic. We propose a novel framework that incorporates an improved outline filling attack to achieve dense sampling of the model's security boundaries. Furthermore, we introduce a ranking regression paradigm that replaces standard regression and trains the proxy model to predict which prompt yields a higher ASR. Experimental results show that our proxy model achieves an accuracy of 91.1 percent in predicting the relative ranking of average long response (ALR), and 69.2 percent in predicting ASR. These findings confirm the predictability and distillability of jailbreak behaviors, and demonstrate the potential of leveraging such distillability to optimize black-box attacks.

Yinlong Bai, Hongxin Zhang, Sheng Zhong et al.

Recent advancements in 3D Gaussian Splatting (3DGS) have made a significant impact on rendering and reconstruction techniques. Current research predominantly focuses on improving rendering performance and reconstruction quality using high-performance desktop GPUs, largely overlooking applications for embedded platforms like micro air vehicles (MAVs). These devices, with their limited computational resources and memory, often face a trade-off between system performance and reconstruction quality. In this paper, we improve existing methods in terms of GPU memory usage while enhancing rendering quality. Specifically, to address redundant 3D Gaussian primitives in SLAM, we propose merging them in voxel space based on geometric similarity. This reduces GPU memory usage without impacting system runtime performance. Furthermore, rendering quality is improved by initializing 3D Gaussian primitives via Patch-Grid (PG) point sampling, enabling more accurate modeling of the entire scene. Quantitative and qualitative evaluations on publicly available datasets demonstrate the effectiveness of our improvements.

Shijun Yang, Xiang Zhang, Wanqing Zhao et al.

Prompt learning facilitates the efficient adaptation of Vision-Language Models (VLMs) to various downstream tasks. However, it faces two significant challenges: (1) inadequate modeling of class embedding distributions for unseen instances, leading to suboptimal generalization on novel classes; (2) prevailing methodologies predominantly confine cross-modal alignment to the final output layer of vision and text encoders, which fundamentally limits their capacity to preserve topological consistency with pre-trained multi-modal embedding spaces. To this end, we introduce MuGCP (Multi-modal Mutual-Guidance Conditional Prompt Learning), a novel paradigm designed for conditional prompt generation. MuGCP leverages Multi-modal Large Language Models (MLLMs) as conditional prompt learners to adaptively generate Semantic Conditional Prompts (SCP) that incorporate rich, fine-grained high-level semantic knowledge for image instances. To ensure effective alignment and interaction across the multi-modal space of Vision-Language Models (VLMs), we introduce the Attention Mutual-Guidance (AMG) module, which facilitates interactions between visual and semantic information. Through mutual guidance, the AMG module generates Visual Conditional Prompts (VCP), enhancing the model's performance in multi-modal tasks. Additionally, we present a Multi-Prompt Fusion (MPF) mechanism that integrates SCP and VCP with contextual prompts, ensuring seamless coordination among the different prompts and enhancing the modeling of class embeddings and instance-specific knowledge. Our MuGCP outperforms existing state-of-the-art methods on 14 different datasets. The code will be made available after publication.

Shucheng Li, Fengyuan Xu, Runchuan Wang et al.

In recent years, phishing scams have become the crime type with the largest money involved on Ethereum, the second-largest blockchain platform. Meanwhile, graph neural network (GNN) has shown promising performance in various node classification tasks. However, for Ethereum transaction data, which could be naturally abstracted to a real-world complex graph, the scarcity of labels and the huge volume of transaction data make it difficult to take advantage of GNN methods. Here in this paper, to address the two challenges, we propose a Self-supervised Incremental deep Graph learning model (SIEGE), for the phishing scam detection problem on Ethereum. In our model, two pretext tasks designed from spatial and temporal perspectives help us effectively learn useful node embedding from the huge amount of unlabelled transaction data. And the incremental paradigm allows us to efficiently handle large-scale transaction data and help the model maintain good performance when the data distribution is drastically changing. We collect transaction records about half a year from Ethereum and our extensive experiments show that our model consistently outperforms strong baselines in both transductive and inductive settings.

Taihang Dong, Sheng Zhong

Discriminative correlation filter (DCF) based trackers have recently achieved excellent performance with great computational efficiency. However, DCF based trackers suffer boundary effects, which result in unstable performance in challenging situations exhibiting fast motion. In this paper, we propose a novel method to mitigate this side-effect in DCF based trackers. We change the search area according to the prediction of target motion. When the object moves fast, broad search area could alleviate boundary effects and reserve the probability of locating the object. When the object moves slowly, narrow search area could prevent effect of useless background information and improve computational efficiency to attain real-time performance. This strategy can impressively soothe boundary effects in situations exhibiting fast motion and motion blur, and it can be used in almost all DCF based trackers. The experiments on OTB benchmark show that the proposed framework improves the performance compared with the baseline trackers.

Taihang Dong, Sheng Zhong

Discriminative correlation filters (DCF) have recently shown excellent performance in visual object tracking area. In this paper, we summarize the methods of updating model filter from discriminative correlation filter (DCF) based tracking algorithms and analyzes similarities and differences among these methods. We deduce the relationship between updating coefficient in high dimension (kernel trick), updating filter in frequency domain and updating filter in spatial domain, and analyze the difference among these different ways. We also analyze the difference between the updating filter directly and updating filter's numerator (object response power) with updating filter's denominator (filter's power). The experiments about comparing different updating methods and visualizing the template filters are used to prove our derivation.

Yi Chang, Luxin Yan, Houzhang Fang et al.

Hyperspectral imaging, providing abundant spatial and spectral information simultaneously, has attracted a lot of interest in recent years. Unfortunately, due to the hardware limitations, the hyperspectral image (HSI) is vulnerable to various degradations, such noises (random noise, HSI denoising), blurs (Gaussian and uniform blur, HSI deblurring), and down-sampled (both spectral and spatial downsample, HSI super-resolution). Previous HSI restoration methods are designed for one specific task only. Besides, most of them start from the 1-D vector or 2-D matrix models and cannot fully exploit the structurally spectral-spatial correlation in 3-D HSI. To overcome these limitations, in this work, we propose a unified low-rank tensor recovery model for comprehensive HSI restoration tasks, in which non-local similarity between spectral-spatial cubic and spectral correlation are simultaneously captured by 3-order tensors. Further, to improve the capability and flexibility, we formulate it as a weighted low-rank tensor recovery (WLRTR) model by treating the singular values differently, and study its analytical solution. We also consider the exclusive stripe noise in HSI as the gross error by extending WLRTR to robust principal component analysis (WLRTR-RPCA). Extensive experiments demonstrate the proposed WLRTR models consistently outperform state-of-the-arts in typical low level vision HSI tasks, including denoising, destriping, deblurring and super-resolution.