arXiv subjects

Thanh-Toan Do

Publications and source records attributed to Thanh-Toan Do.

At least 37 records · Page 2Linked to original sources

arXiv2025

Multiverse Through Deepfakes: The MultiFakeVerse Dataset of Person-Centric Visual and Conceptual Manipulations

The rapid advancement of GenAI technology over the past few years has significantly contributed towards highly realistic deepfake content generation. Despite ongoing efforts, the research community still lacks a large-scale and reasoning capability driven deepfake benchmark dataset specifically tailored for person-centric object, context and scene manipulations. In this paper, we address this gap by introducing MultiFakeVerse, a large scale person-centric deepfake dataset, comprising 845,286 images generated through manipulation suggestions and image manipulations both derived from vision-language models (VLM). The VLM instructions were specifically targeted towards modifications to individuals or contextual elements of a scene that influence human perception of importance, intent, or narrative. This VLM-driven approach enables semantic, context-aware alterations such as modifying actions, scenes, and human-object interactions rather than synthetic or low-level identity swaps and region-specific edits that are common in existing datasets. Our experiments reveal that current state-of-the-art deepfake detection models and human observers struggle to detect these subtle yet meaningful manipulations. The code and dataset are available on \href{https://github.com/Parul-Gupta/MultiFakeVerse}{GitHub}.

cs.MM↗

arXiv2025

Preserving Clusters in Prompt Learning for Unsupervised Domain Adaptation

Recent approaches leveraging multi-modal pre-trained models like CLIP for Unsupervised Domain Adaptation (UDA) have shown significant promise in bridging domain gaps and improving generalization by utilizing rich semantic knowledge and robust visual representations learned through extensive pre-training on diverse image-text datasets. While these methods achieve state-of-the-art performance across benchmarks, much of the improvement stems from base pseudo-labels (CLIP zero-shot predictions) and self-training mechanisms. Thus, the training mechanism exhibits a key limitation wherein the visual embedding distribution in target domains can deviate from the visual embedding distribution in the pre-trained model, leading to misguided signals from class descriptions. This work introduces a fresh solution to reinforce these pseudo-labels and facilitate target-prompt learning, by exploiting the geometry of visual and text embeddings - an aspect that is overlooked by existing methods. We first propose to directly leverage the reference predictions (from source prompts) based on the relationship between source and target visual embeddings. We later show that there is a strong clustering behavior observed between visual and text embeddings in pre-trained multi-modal models. Building on optimal transport theory, we transform this insight into a novel strategy to enforce the clustering property in text embeddings, further enhancing the alignment in the target domain. Our experiments and ablation studies validate the effectiveness of the proposed approach, demonstrating superior performance and improved quality of target prompts in terms of representation.

cs.CV↗

arXiv2025

Vision-Based Assistive Technologies for People with Cerebral Visual Impairment: A Review and Focus Study

Over the past decade, considerable research has investigated Vision-Based Assistive Technologies (VBAT) to support people with vision impairments to understand and interact with their immediate environment using machine learning, computer vision, image enhancement, and/or augmented/virtual reality. However, this has almost totally overlooked a growing demographic: people with Cerebral Visual Impairment (CVI). Unlike ocular vision impairments, CVI arises from damage to the brain's visual processing centres. Through a scoping review, this paper reveals a significant research gap in addressing the needs of this demographic. Three focus studies involving 7 participants with CVI explored the challenges, current strategies, and opportunities for VBAT. We also discussed the assistive technology needs of people with CVI compared with ocular low vision. Our findings highlight the opportunity for the Human-Computer Interaction and Assistive Technologies research community to explore and address this underrepresented domain, thereby enhancing the quality of life for people with CVI.

cs.HC↗

arXiv2025

Broadening Our View: Assistive Technology for Cerebral Visual Impairment

Over the past decade, considerable research has been directed towards assistive technologies to support people with vision impairments using machine learning, computer vision, image enhancement, and/or augmented/virtual reality. However, this has almost totally overlooked a growing demographic: people with Cerebral Visual Impairment (CVI). Unlike Ocular Vision Impairments (OVI), CVI arises from damage to the brain's visual processing centres. This paper introduces CVI and reveals a wide research gap in addressing the needs of this demographic. Through a scoping review, we identified 14 papers at the intersection of these technologies and CVI. Of these, only three papers described assistive technologies focused on people living with CVI, with the others focusing on diagnosis, understanding, simulation or rehabilitation. Our findings highlight the opportunity for the Human-Computer Interaction and Assistive Technologies research community to explore and address this underrepresented domain, thereby enhancing the quality of life for people with CVI.

cs.HC↗

arXiv2025

What do Blind and Low-Vision People Really Want from Assistive Smart Devices? Comparison of the Literature with a Focus Study

Over the last decade there has been considerable research into how artificial intelligence (AI), specifically computer vision, can assist people who are blind or have low-vision (BLV) to understand their environment. However, there has been almost no research into whether the tasks (object detection, image captioning, text recognition etc.) and devices (smartphones, smart-glasses etc.) investigated by researchers align with the needs and preferences of BLV people. We identified 646 studies published in the last two and a half years that have investigated such assistive AI techniques. We analysed these papers to determine the task, device and participation by BLV individuals. We then interviewed 24 BLV people and asked for their top five AI-based applications and to rank the applications found in the literature. We found only a weak positive correlation between BLV participants' perceived importance of tasks and researchers' focus and that participants prefer conversational agent interface and head-mounted devices.

cs.HC↗

arXiv2025

Self-supervised Learning for Acoustic Few-Shot Classification

Labelled data are limited and self-supervised learning is one of the most important approaches for reducing labelling requirements. While it has been extensively explored in the image domain, it has so far not received the same amount of attention in the acoustic domain. Yet, reducing labelling is a key requirement for many acoustic applications. Specifically in bioacoustic, there are rarely sufficient labels for fully supervised learning available. This has led to the widespread use of acoustic recognisers that have been pre-trained on unrelated data for bioacoustic tasks. We posit that training on the actual task data and combining self-supervised pre-training with few-shot classification is a superior approach that has the ability to deliver high accuracy even when only a few labels are available. To this end, we introduce and evaluate a new architecture that combines CNN-based preprocessing with feature extraction based on state space models (SSMs). This combination is motivated by the fact that CNN-based networks alone struggle to capture temporal information effectively, which is crucial for classifying acoustic signals. SSMs, specifically S4 and Mamba, on the other hand, have been shown to have an excellent ability to capture long-range dependencies in sequence data. We pre-train this architecture using contrastive learning on the actual task data and subsequent fine-tuning with an extremely small amount of labelled data. We evaluate the performance of this proposed architecture for ($n$-shot, $n$-class) classification on standard benchmarks as well as real-world data. Our evaluation shows that it outperforms state-of-the-art architectures on the few-shot classification problem.

cs.SD↗

arXiv2025

RoTipBot: Robotic Handling of Thin and Flexible Objects using Rotatable Tactile Sensors

This paper introduces RoTipBot, a novel robotic system for handling thin, flexible objects. Different from previous works that are limited to singulating them using suction cups or soft grippers, RoTipBot can count multiple layers and then grasp them simultaneously in a single grasp closure. Specifically, we first develop a vision-based tactile sensor named RoTip that can rotate and sense contact information around its tip. Equipped with two RoTip sensors, RoTipBot rolls and feeds multiple layers of thin, flexible objects into the centre between its fingers, enabling effective grasping. Moreover, we design a tactile-based grasping strategy that uses RoTip's sensing ability to ensure both fingers maintain secure contact with the object while accurately counting the number of fed objects. Extensive experiments demonstrate the efficacy of the RoTip sensor and the RoTipBot approach. The results show that RoTipBot not only achieves a higher success rate but also grasps and counts multiple layers simultaneously -- capabilities not possible with previous methods. Furthermore, RoTipBot operates up to three times faster than state-of-the-art methods. The success of RoTipBot paves the way for future research in object manipulation using mobilised tactile sensors. All the materials used in this paper are available at https://sites.google.com/view/rotipbot.

cs.RO↗

arXiv2025

Optimizing Specific and Shared Parameters for Efficient Parameter Tuning

Foundation models, with a vast number of parameters and pretraining on massive datasets, achieve state-of-the-art performance across various applications. However, efficiently adapting them to downstream tasks with minimal computational overhead remains a challenge. Parameter-Efficient Transfer Learning (PETL) addresses this by fine-tuning only a small subset of parameters while preserving pre-trained knowledge. In this paper, we propose SaS, a novel PETL method that effectively mitigates distributional shifts during fine-tuning. SaS integrates (1) a shared module that captures common statistical characteristics across layers using low-rank projections and (2) a layer-specific module that employs hypernetworks to generate tailored parameters for each layer. This dual design ensures an optimal balance between performance and parameter efficiency while introducing less than 0.05% additional parameters, making it significantly more compact than existing methods. Extensive experiments on diverse downstream tasks, few-shot settings and domain generalization demonstrate that SaS significantly enhances performance while maintaining superior parameter efficiency compared to existing methods, highlighting the importance of capturing both shared and layer-specific information in transfer learning. Code and data are available at https://anonymous.4open.science/r/SaS-PETL-3565.

cs.LG↗

arXiv2025

Enhancing Dataset Distillation via Non-Critical Region Refinement

Dataset distillation has become a popular method for compressing large datasets into smaller, more efficient representations while preserving critical information for model training. Data features are broadly categorized into two types: instance-specific features, which capture unique, fine-grained details of individual examples, and class-general features, which represent shared, broad patterns across a class. However, previous approaches often struggle to balance these features-some focus solely on class-general patterns, neglecting finer instance details, while others prioritize instance-specific features, overlooking the shared characteristics essential for class-level understanding. In this paper, we introduce the Non-Critical Region Refinement Dataset Distillation (NRR-DD) method, which preserves instance-specific details and fine-grained regions in synthetic data while enriching non-critical regions with class-general information. This approach enables models to leverage all pixel information, capturing both feature types and enhancing overall performance. Additionally, we present Distance-Based Representative (DBR) knowledge transfer, which eliminates the need for soft labels in training by relying on the distance between synthetic data predictions and one-hot encoded labels. Experimental results show that NRR-DD achieves state-of-the-art performance on both small- and large-scale datasets. Furthermore, by storing only two distances per instance, our method delivers comparable results across various settings. The code is available at https://github.com/tmtuan1307/NRR-DD.

cs.CV↗

arXiv2025

Multi-Perspective Data Augmentation for Few-shot Object Detection

Recent few-shot object detection (FSOD) methods have focused on augmenting synthetic samples for novel classes, show promising results to the rise of diffusion models. However, the diversity of such datasets is often limited in representativeness because they lack awareness of typical and hard samples, especially in the context of foreground and background relationships. To tackle this issue, we propose a Multi-Perspective Data Augmentation (MPAD) framework. In terms of foreground-foreground relationships, we propose in-context learning for object synthesis (ICOS) with bounding box adjustments to enhance the detail and spatial information of synthetic samples. Inspired by the large margin principle, support samples play a vital role in defining class boundaries. Therefore, we design a Harmonic Prompt Aggregation Scheduler (HPAS) to mix prompt embeddings at each time step of the generation process in diffusion models, producing hard novel samples. For foreground-background relationships, we introduce a Background Proposal method (BAP) to sample typical and hard backgrounds. Extensive experiments on multiple FSOD benchmarks demonstrate the effectiveness of our approach. Our framework significantly outperforms traditional methods, achieving an average increase of $17.5\%$ in nAP50 over the baseline on PASCAL VOC. Code is available at https://github.com/nvakhoa/MPAD.

cs.CV↗

arXiv2025

Why Domain Generalization Fail? A View of Necessity and Sufficiency

Despite a strong theoretical foundation, empirical experiments reveal that existing domain generalization (DG) algorithms often fail to consistently outperform the ERM baseline. We argue that this issue arises because most DG studies focus on establishing theoretical guarantees for generalization under unrealistic assumptions, such as the availability of sufficient, diverse (or even infinite) domains or access to target domain knowledge. As a result, the extent to which domain generalization is achievable in scenarios with limited domains remains largely unexplored. This paper seeks to address this gap by examining generalization through the lens of the conditions necessary for its existence and learnability. Specifically, we systematically establish a set of necessary and sufficient conditions for generalization. Our analysis highlights that existing DG methods primarily act as regularization mechanisms focused on satisfying sufficient conditions, while often neglecting necessary ones. However, sufficient conditions cannot be verified in settings with limited training domains. In such cases, regularization targeting sufficient conditions aims to maximize the likelihood of generalization, whereas regularization targeting necessary conditions ensures its existence. Using this analysis, we reveal the shortcomings of existing DG algorithms by showing that, while they promote sufficient conditions, they inadvertently violate necessary conditions. To validate our theoretical insights, we propose a practical method that promotes the sufficient condition while maintaining the necessary conditions through a novel subspace representation alignment strategy. This approach highlights the advantages of preserving the necessary conditions on well-established DG benchmarks.

cs.LG↗

arXiv2025

AEON: Adaptive Estimation of Instance-Dependent In-Distribution and Out-of-Distribution Label Noise for Robust Learning

Robust training with noisy labels is a critical challenge in image classification, offering the potential to reduce reliance on costly clean-label datasets. Real-world datasets often contain a mix of in-distribution (ID) and out-of-distribution (OOD) instance-dependent label noise, a challenge that is rarely addressed simultaneously by existing methods and is further compounded by the lack of comprehensive benchmarking datasets. Furthermore, even though current noisy-label learning approaches attempt to find noisy-label samples during training, these methods do not aim to estimate ID and OOD noise rates to promote their effectiveness in the selection of such noisy-label samples, and they are often represented by inefficient multi-stage learning algorithms. We propose the Adaptive Estimation of Instance-Dependent In-Distribution and Out-of-Distribution Label Noise (AEON) approach to address these research gaps. AEON is an efficient one-stage noisy-label learning methodology that dynamically estimates instance-dependent ID and OOD label noise rates to enhance robustness to complex noise settings. Additionally, we introduce a new benchmark reflecting real-world ID and OOD noise scenarios. Experiments demonstrate that AEON achieves state-of-the-art performance on both synthetic and real-world datasets

cs.CV↗

arXiv2024

Coverage-Constrained Human-AI Cooperation with Multiple Experts

Human-AI cooperative classification (HAI-CC) approaches aim to develop hybrid intelligent systems that enhance decision-making in various high-stakes real-world scenarios by leveraging both human expertise and AI capabilities. Current HAI-CC methods primarily focus on learning-to-defer (L2D), where decisions are deferred to human experts, and learning-to-complement (L2C), where AI and human experts make predictions cooperatively. However, a notable research gap remains in effectively exploring both L2D and L2C under diverse expert knowledge to improve decision-making, particularly when constrained by the cooperation cost required to achieve a target probability for AI-only selection (i.e., coverage). In this paper, we address this research gap by proposing the Coverage-constrained Learning to Defer and Complement with Specific Experts (CL2DC) method. CL2DC makes final decisions through either AI prediction alone or by deferring to or complementing a specific expert, depending on the input data. Furthermore, we propose a coverage-constrained optimisation to control the cooperation cost, ensuring it approximates a target probability for AI-only selection. This approach enables an effective assessment of system performance within a specified budget. Also, CL2DC is designed to address scenarios where training sets contain multiple noisy-label annotations without any clean-label references. Comprehensive evaluations on both synthetic and real-world datasets demonstrate that CL2DC achieves superior performance compared to state-of-the-art HAI-CC methods.

cs.LG↗

arXiv2024

CamoFA: A Learnable Fourier-based Augmentation for Camouflage Segmentation

Camouflaged object detection (COD) and camouflaged instance segmentation (CIS) aim to recognize and segment objects that are blended into their surroundings, respectively. While several deep neural network models have been proposed to tackle those tasks, augmentation methods for COD and CIS have not been thoroughly explored. Augmentation strategies can help improve models' performance by increasing the size and diversity of the training data and exposing the model to a wider range of variations in the data. Besides, we aim to automatically learn transformations that help to reveal the underlying structure of camouflaged objects and allow the model to learn to better identify and segment camouflaged objects. To achieve this, we propose a learnable augmentation method in the frequency domain for COD and CIS via the Fourier transform approach, dubbed CamoFA. Our method leverages a conditional generative adversarial network and cross-attention mechanism to generate a reference image and an adaptive hybrid swapping with parameters to mix the low-frequency component of the reference image and the high-frequency component of the input image. This approach aims to make camouflaged objects more visible for detection and segmentation models. Without bells and whistles, our proposed augmentation method boosts the performance of camouflaged object detectors and instance segmenters by large margins.

cs.CV↗

arXiv2024

Fair Distillation: Teaching Fairness from Biased Teachers in Medical Imaging

Deep learning has achieved remarkable success in image classification and segmentation tasks. However, fairness concerns persist, as models often exhibit biases that disproportionately affect demographic groups defined by sensitive attributes such as race, gender, or age. Existing bias-mitigation techniques, including Subgroup Re-balancing, Adversarial Training, and Domain Generalization, aim to balance accuracy across demographic groups, but often fail to simultaneously improve overall accuracy, group-specific accuracy, and fairness due to conflicts among these interdependent objectives. We propose the Fair Distillation (FairDi) method, a novel fairness approach that decomposes these objectives by leveraging biased ``teacher'' models, each optimized for a specific demographic group. These teacher models then guide the training of a unified ``student'' model, which distills their knowledge to maximize overall and group-specific accuracies, while minimizing inter-group disparities. Experiments on medical imaging datasets show that FairDi achieves significant gains in both overall and group-specific accuracy, along with improved fairness, compared to existing methods. FairDi is adaptable to various medical tasks, such as classification and segmentation, and provides an effective solution for equitable model performance.

cs.CV↗

arXiv2024

Agnostic Sharpness-Aware Minimization

Sharpness-aware minimization (SAM) has been instrumental in improving deep neural network training by minimizing both the training loss and the sharpness of the loss landscape, leading the model into flatter minima that are associated with better generalization properties. In another aspect, Model-Agnostic Meta-Learning (MAML) is a framework designed to improve the adaptability of models. MAML optimizes a set of meta-models that are specifically tailored for quick adaptation to multiple tasks with minimal fine-tuning steps and can generalize well with limited data. In this work, we explore the connection between SAM and MAML in enhancing model generalization. We introduce Agnostic-SAM, a novel approach that combines the principles of both SAM and MAML. Agnostic-SAM adapts the core idea of SAM by optimizing the model toward wider local minima using training data, while concurrently maintaining low loss values on validation data. By doing so, it seeks flatter minima that are not only robust to small perturbations but also less vulnerable to data distributional shift problems. Our experimental results demonstrate that Agnostic-SAM significantly improves generalization over baselines across a range of datasets and under challenging conditions such as noisy labels or data limitation.

cs.LG↗

arXiv2024

Connective Viewpoints of Signal-to-Noise Diffusion Models

Diffusion models (DM) have become fundamental components of generative models, excelling across various domains such as image creation, audio generation, and complex data interpolation. Signal-to-Noise diffusion models constitute a diverse family covering most state-of-the-art diffusion models. While there have been several attempts to study Signal-to-Noise (S2N) diffusion models from various perspectives, there remains a need for a comprehensive study connecting different viewpoints and exploring new perspectives. In this study, we offer a comprehensive perspective on noise schedulers, examining their role through the lens of the signal-to-noise ratio (SNR) and its connections to information theory. Building upon this framework, we have developed a generalized backward equation to enhance the performance of the inference process.

cs.CV↗

arXiv2024

The Art of Camouflage: Few-Shot Learning for Animal Detection and Segmentation

Camouflaged object detection and segmentation is a new and challenging research topic in computer vision. There is a serious issue of lacking data on concealed objects such as camouflaged animals in natural scenes. In this paper, we address the problem of few-shot learning for camouflaged object detection and segmentation. To this end, we first collect a new dataset, CAMO-FS, for the benchmark. As camouflaged instances are challenging to recognize due to their similarity compared to the surroundings, we guide our models to obtain camouflaged features that highly distinguish the instances from the background. In this work, we propose FS-CDIS, a framework to efficiently detect and segment camouflaged instances via two loss functions contributing to the training process. Firstly, the instance triplet loss with the characteristic of differentiating the anchor, which is the mean of all camouflaged foreground points, and the background points are employed to work at the instance level. Secondly, to consolidate the generalization at the class level, we present instance memory storage with the scope of storing camouflaged features of the same category, allowing the model to capture further class-level information during the learning process. The extensive experiments demonstrated that our proposed method achieves state-of-the-art performance on the newly collected dataset. Code is available at https://github.com/danhntd/FS-CDIS.

cs.CV↗