Search Results for author:
Found 62 papers, 15 papers with code
FedCCL: Federated Dual-Clustered Feature Contrast Under Domain Heterogeneity
First, we employ clustering on the local representations of each client, aiming to capture intra-class information based on these local clusters at a high level of granularity.
Logit Calibration and Feature Contrast for Robust Federated Learning on Non-IID Data
Federated learning (FL) is a privacy-preserving distributed framework for collaborative model training on devices in edge networks.
Towards Understanding Dual BN In Hybrid Adversarial Training
There is a growing concern about applying batch normalization (BN) in adversarial training (AT), especially when the model is trained on both adversarial samples and clean samples (termed Hybrid-AT).
Sora as an AGI World Model? A Complete Survey on Text-to-Video Generation
Text-to-video generation marks a significant frontier in the rapidly evolving domain of generative AI, integrating advancements in text-to-image synthesis, video captioning, and text-guided editing.
Towards Robust Federated Learning via Logits Calibration on Non-IID Data
Meanwhile, the non-independent and identically distributed (non-IID) challenge of data distribution between edge devices can further degrade the performance of models.
MobileSAMv2: Faster Segment Anything to Everything
The efficiency bottleneck of SegEvery with SAM, however, lies in its mask decoder because it needs to first generate numerous masks with redundant grid-search prompts and then perform filtering to obtain the final valid masks.
Single Image Reflection Removal with Reflection Intensity Prior Knowledge
In this paper, we propose a general reflection intensity prior that captures the intensity of the reflection phenomenon and demonstrate its effectiveness.
Federated Learning with Diffusion Models for Privacy-Sensitive Vision Tasks
We conduct experiments on various FL scenarios, and our findings demonstrate that federated diffusion models have great potential to deliver vision services to privacy-sensitive domains.
Segment Anything Meets Universal Adversarial Perturbation
On top of the ablation study to understand various components in our proposed method, we shed light on the roles of positive and negative samples in making the generated UAP effective for attacking SAM.
Black-box Targeted Adversarial Attack on Segment Anything (SAM)
The task of TAA on SAM has been realized in a recent arXiv work in the white-box setup by assuming access to prompt and model, which is thus less practical.
DimCL: Dimensional Contrastive Learning For Improving Self-Supervised Learning
Self-supervised learning (SSL) has gained remarkable success, for which contrastive learning (CL) plays a key role.
Toward a Deeper Understanding: RetNet Viewed through Convolution
A straightforward way to locally adapt the self-attention matrix can be realized by an element-wise learnable weight mask (ELM), for which our preliminary results show promising results.
FedMEKT: Distillation-based Embedding Knowledge Transfer for Multimodal Federated Learning
Bringing this concept into a system, we develop a distillation-based multimodal embedding knowledge transfer mechanism, namely FedMEKT, which allows the server and clients to exchange the joint knowledge of their learning models extracted from a small multimodal proxy dataset.
Internal-External Boundary Attention Fusion for Glass Surface Segmentation
Inspired by prior semantic segmentation approaches with challenging image types such as X-ray or CT scans, we propose separated internal-external boundary attention modules that individually learn and selectively integrate visual characteristics of the inside and outside region of glass surface from a single color image.
Faster Segment Anything: Towards Lightweight SAM for Mobile Applications
Concretely, we distill the knowledge from the heavy image encoder (ViT-H in the original SAM) to a lightweight image encoder, which can be automatically compatible with the mask decoder in the original SAM.
Robustness of Segment Anything Model (SAM) for Autonomous Driving in Adverse Weather Conditions
Given its impressive performance, there is a strong desire to apply SAM in autonomous driving to improve the performance of vision tasks, particularly in challenging scenarios such as driving under adverse weather conditions.
Robustness of SAM: Segment Anything Under Corruptions and Beyond
Following by interpreting the effects of synthetic corruption as style changes, we proceed to conduct a comprehensive evaluation for its robustness against 15 types of common corruption.
Segment Anything Meets Semantic Communication
In light of the diminishing returns of traditional methods for enhancing transmission rates, the domain of semantic communication presents promising new frontiers.
Understanding Segment Anything Model: SAM is Biased Towards Texture Rather than Shape
Different from label-oriented recognition tasks, the SAM is trained to predict a mask for covering the object shape based on a promt.
A Survey on Segment Anything Model (SAM): Vision Foundation Model Meets Prompt Engineering
This is an ongoing project and we intend to update the manuscript on a regular basis.
Generative AI meets 3D: A Survey on Text-to-3D in AIGC Era
AI generated content) has made remarkable progress in the past few years, among which text-guided content generation is the most practical one since it enables the interaction between human instruction and AIGC.
When ChatGPT for Computer Vision Will Come? From 2D to 3D
On top of that, this work presents an outlook on the development of AIGC in 3D from the data perspective.
Attack-SAM: Towards Attacking Segment Anything Model With Adversarial Examples
Beyond the basic goal of mask removal, we further investigate and find that it is possible to generate any desired mask by the adversarial attack.
Segment Anything Model (SAM) Meets Glass: Mirror and Transparent Objects Cannot Be Easily Detected
Meta AI Research has recently released SAM (Segment Anything Model) which is trained on a large segmentation dataset of over 1 billion masks.
One Small Step for Generative AI, One Giant Leap for AGI: A Complete Survey on ChatGPT in AIGC Era
Overall, this work is the first to survey ChatGPT with a comprehensive review of its underlying technology, applications, and challenges.
A Survey on Graph Diffusion Models: Generative AI in Science for Molecule, Protein and Material
Diffusion models have become a new SOTA generative modeling method in various fields, for which there are multiple survey works that provide an overall survey.
MP-FedCL: Multiprototype Federated Contrastive Learning for Edge Intelligence
The existing single prototype-based strategy represents a class by using the mean of the feature space.
A Survey on Audio Diffusion Models: Text To Speech Synthesis and Enhancement in Generative AI
This work conducts a survey on audio diffusion model, which is complementary to existing surveys that either lack the recent progress of diffusion-based speech synthesis or highlight an overall picture of applying diffusion model in multiple fields.
A Complete Survey on Generative AI (AIGC): Is ChatGPT from GPT-4 to GPT-5 All You Need?
As ChatGPT goes viral, generative AI (AIGC, a. k. a AI-generated content) has made headlines everywhere because of its ability to analyze and create text, images, and beyond.
Text-to-image Diffusion Models in Generative AI: A Survey
This survey reviews text-to-image diffusion models in the context that diffusion models have emerged to be popular for a wide range of generative tasks.
Test-time Adaptation in the Dynamic World with Compound Domain Knowledge Management
In addition, to prevent overfitting of the TTA model, we devise novel regularization which modulates the adaptation rates using domain-similarity between the source and the current target domain.
On the Pros and Cons of Momentum Encoder in Self-Supervised Visual Representation Learning
Exponential Moving Average (EMA or momentum) is widely used in modern self-supervised learning (SSL) approaches, such as MoCo, for enhancing performance.
A Survey on Masked Autoencoder for Self-supervised Learning in Vision and Beyond
Masked autoencoders are scalable vision learners, as the title of MAE \cite{he2022masked}, which suggests that self-supervised learning (SSL) in vision might undertake a similar trajectory as in NLP.
Decoupled Adversarial Contrastive Learning for Self-supervised Adversarial Robustness
Adversarial training (AT) for robust representation learning and self-supervised learning (SSL) for unsupervised representation learning are two active research fields.
Investigating Top-$k$ White-Box and Transferable Black-box Attack
It is widely reported that stronger I-FGSM transfers worse than simple FGSM, leading to a popular belief that transferability is at odds with the white-box attack strength.
How Does SimSiam Avoid Collapse Without Negative Samples? A Unified Understanding with Self-supervised Contrastive Learning
This yields a unified perspective on how negative samples and SimSiam alleviate collapse.
Dual Temperature Helps Contrastive Learning Without Many Negative Samples: Towards Understanding and Simplifying MoCo
Contrastive learning (CL) is widely known to require many negative samples, 65536 in MoCo for instance, for which the performance of a dictionary-free framework is often inferior because the negative sample size (NSS) is limited by its mini-batch size (MBS).
Practical No-box Adversarial Attacks with Training-free Hybrid Image Transformation
In this paper, we move a step forward and show the existence of a \textbf{training-free} adversarial perturbation under the no-box threat model, which can be successfully used to attack different DNNs in real-time.
Fast Adversarial Training with Noise Augmentation: A Unified Perspective on RandStart and GradAlign
The former works only for a relatively small perturbation 8/255 with the l_\infty constraint, and GradAlign improves it by extending the perturbation size to 16/255 (with the l_\infty constraint) but at the cost of being 3 to 4 times slower.
Investigating Top-k White-Box and Transferable Black-Box Attack
It is widely reported that stronger I-FGSM transfers worse than simple FGSM, leading to a popular belief that transferability is at odds with the white-box attack strength.
Unrestricted Adversarial Attacks on ImageNet Competition
Many works have investigated the adversarial attacks or defenses under the settings where a bounded and imperceptible perturbation can be added to the input.
Adversarial Robustness Comparison of Vision Transformer and MLP-Mixer to CNNs
Thus, it is critical for the community to know whether the newly proposed ViT and MLP-Mixer are also vulnerable to adversarial attacks.
Early Stop And Adversarial Training Yield Better surrogate Model: Very Non-Robust Features Harm Adversarial Transferability
The transferability of adversarial examples (AE); known as adversarial transferability, has attracted significant attention because it can be exploited for TransferableBlack-box Attacks (TBA).
How Does SimSiam Avoid Collapse Without Negative Samples? Towards a Unified Understanding of Progress in SSL
Towards avoiding collapse in self-supervised learning (SSL), contrastive loss is widely used but often requires a large number of negative samples.
Universal Adversarial Head: Practical Protection against Video Data Leakage
We propose $universal \ adversarial \ head$ (UAH), which crafts adversarial query videos by prepending the original videos with a sequence of adversarial frames to perturb the normal hash codes in the Hamming space.
Restoration of Video Frames from a Single Blurred Image with Motion Understanding
We formulate video restoration from a single blurred image as an inverse problem by setting clean image sequence and their respective motion as latent factors, and the blurred image as an observation.
Universal Adversarial Training with Class-Wise Perturbations
The SOTA universal adversarial training (UAT) method optimizes a single perturbation for all training samples in the mini-batch.
A Survey On Universal Adversarial Attack
The intriguing phenomenon of adversarial examples has attracted significant attention in machine learning and what might be more surprising to the community is the existence of universal adversarial perturbations (UAPs), i. e. a single perturbation to fool the target DNN for most images.
A Brief Survey on Deep Learning Based Data Hiding
Data hiding is the art of concealing messages with limited perceptual changes.
Universal Adversarial Perturbations Through the Lens of Deep Steganography: Towards A Fourier Perspective
We perform task-specific and joint analysis and reveal that (a) frequency is a key factor that influences their performance based on the proposed entropy metric for quantifying the frequency distribution; (b) their success can be attributed to a DNN being highly sensitive to high-frequency content.
Data-Free Universal Adversarial Perturbation and Black-Box Attack
For a more practical universal attack, our investigation of untargeted UAP focuses on alleviating the dependence on the original training samples, from removing the need for sample labels to limiting the sample size.
Towards Robust Data Hiding Against (JPEG) Compression: A Pseudo-Differentiable Deep Learning Approach
Recently, deep learning has shown large success in data hiding, while non-differentiability of JPEG makes it challenging to train a deep pipeline for improving robustness against lossy compression.
UDH: Universal Deep Hiding for Steganography, Watermarking, and Light Field Messaging
This is the first work demonstrating the success of (DNN-based) hiding a full image for watermarking and LFM.
Robustness May Be at Odds with Fairness: An Empirical Study on Class-wise Accuracy
Adversarial training is the most widely used technique for improving adversarial robustness to strong white-box attacks.
CD-UAP: Class Discriminative Universal Adversarial Perturbation
Since the proposed attack generates a universal adversarial perturbation that is discriminative to targeted and non-targeted classes, we term it class discriminative universal adversarial perturbation (CD-UAP).
Double Targeted Universal Adversarial Perturbations
This universal perturbation attacks one targeted source class to sink class, while having a limited adversarial effect on other non-targeted source classes, for avoiding raising suspicions.
Batch Normalization Increases Adversarial Vulnerability and Decreases Adversarial Transferability: A Non-Robust Feature Perspective
This work attempts to understand the impact of BN on DNNs from a non-robust feature perspective.
Revisiting Batch Normalization for Improving Corruption Robustness
We find that simply estimating and adapting the BN statistics on a few (32 for instance) representation samples, without retraining the model, improves the corruption robustness by a large margin on several benchmark datasets with a wide range of model architectures.
Data from Model: Extracting Data from Non-robust and Robust Models
We repeat the process of Data to Model (DtM) and Data from Model (DfM) in sequence and explore the loss of feature mapping information by measuring the accuracy drop on the original validation dataset.
Understanding Adversarial Examples from the Mutual Influence of Images and Perturbations
We utilize this vector representation to understand adversarial examples by disentangling the clean images and adversarial perturbations, and analyze their influence on each other.
