Auxiliary Prompt Tuning Of Vision‑Language Models For Few‑Shot Out‑Of‑Distribution Detection,
2025
Singapore Management University
Auxiliary Prompt Tuning Of Vision‑Language Models For Few‑Shot Out‑Of‑Distribution Detection, Wenjun Miao, Guansong Pang, Zihan Wang, Jin Zheng, Xiao Bai
Research Collection School Of Computing and Information Systems
Recent advancements in CLIP-based out-of-distribution (OOD) detection have shown promising results via regularization on prompt tuning, leveraging background features extracted from a few in-distribution (ID) samples as proxies for OOD features.However, these methods suffer from an inherent limitation: a lack of diversity in the extracted OOD features from the few-shot ID data.To address this issue, we propose to leverage external datasets as auxiliary outlier data (i.e., pseudo OOD samples) to extract rich, diverse OOD features, with the features from not only background regions but also foreground object regions, thereby supporting more discriminative prompt tuning for OOD detection. We further introduce …
Viewsrd: 3d Visual Grounding Via Structured Multi-View Decomposition,
2025
Singapore Management University
Viewsrd: 3d Visual Grounding Via Structured Multi-View Decomposition, Ronggang Huang, Haoxin Yang, Yan Cai, Xuemiao Xu, Huaidong Zhang, Shengfeng He
Research Collection School Of Computing and Information Systems
3Dvisual grounding aims to identify and localize objects in a 3Dspacebasedontextualdescriptions. However, existing methods struggle with disentangling targets from anchors in complex multi-anchor queries and resolving inconsisten cies in spatial descriptions caused by perspective variations. To tackle these challenges, we propose ViewSRD, a frame work that formulates 3D visual grounding as a structured multi-view decomposition process. First, the Simple Rela tion Decoupling (SRD) module restructures complex multi anchor queries into a set of targeted single-anchor state ments, generating a structured set of perspective-aware de scriptions that clarify positional relationships. These de composed representations serve as the foundation for the Multi-view …
A Comprehensive Review Of Financial Knowledge Graphs,
2025
Singapore Management University
A Comprehensive Review Of Financial Knowledge Graphs, Jeyaraman Brindha Priyadarshini, Bing Tian Dai, Yuan Fang
Research Collection School Of Computing and Information Systems
Knowledge Graphs (KGs) are increasingly used in finance to manage complex, interconnected data and support advanced analytics. This survey provides an overview of how KGs are applied across various financial areas, such as fraud detection, credit risk assessment, anti-money laundering, and regulatory compliance. We examine key techniques for building and using KGs in finance, including graph construction, embedding methods, and machine learning models. The survey also discusses challenges specific to finance, like handling private data, ensuring interpretability, and managing real-time data. Additionally, we explore the emerging combination of KGs with large language models and generative AI, which offers new possibilities …
Seeing 3d Through 2d Lenses: 3d Few-Shot Class-Incremental Learning Via Cross-Modal Geometric Rectification,
2025
Singapore Management University
Seeing 3d Through 2d Lenses: 3d Few-Shot Class-Incremental Learning Via Cross-Modal Geometric Rectification, Tuo Xiang, Xuemiao Xu, Bangzhen Liu, Jinyi Li, Yong Li, Shengfeng He
Research Collection School Of Computing and Information Systems
The rapid growth of 3D digital content necessitates expandable recognition systems for open-world scenarios. However, existing 3D class-incremental learning methods struggle under extreme data scarcity due to geometric misalignment and texture bias. While recent approaches integrate 3D data with 2D foundation models (e.g., CLIP), they suffer from semantic blurring caused by texture-biased projections and indiscriminate fusion of geometric-textural cues, leading to unstable decision prototypes and catastrophic forgetting. To address these issues, we propose Cross-Modal Geometric Rectification (CMGR), a framework that enhances 3D geometric fidelity by leveraging CLIP’s hierarchical spatial semantics. Specifically, we introduce a Structure-Aware Geometric Rectification module that hierarchically …
Omnivton: Training-Free Universal Virtual Try-On,
2025
Singapore Management University
Omnivton: Training-Free Universal Virtual Try-On, Zhaotong Yang, Yuhui Li, Shengfeng He, Xinzhe Li, Yangyang Xu, Junyu Dong, Yong Du
Research Collection School Of Computing and Information Systems
Image-based Virtual Try-On (VTON) techniques rely on either supervised in-shop approaches, which ensure high fidelity but struggle with cross-domain generalization, or unsupervised in-the-wild methods, which improve adaptability but remain constrained by data biases and limited universality. A unified, training-free solution that works across both scenarios remains an open challenge. We propose OmniVTON, the first training-free universal VTON framework that decouples garment and pose conditioning to achieve both texture fidelity and pose consistency across diverse settings. To preserve garment details, we introduce a garment prior generation mechanism that aligns clothing with the body, followed by continuous boundary stitching technique to achieve …
Fcad: Feature-Coupled Anisotropic Diffusion For Continuous Graph Learning,
2025
Singapore Management University
Fcad: Feature-Coupled Anisotropic Diffusion For Continuous Graph Learning, Amitoz Azad, Zhiyuan Zhang
Research Collection School Of Computing and Information Systems
In this work, we propose a novel continuous graph neural network called FCAD (Feature-Coupled Anisotropic Diffusion) for the task of node classification on graphs. Our approach is motivated by the success of feature-coupled anisotropic diffusion PDEs in multivalued image restoration. Our method introduces a total variation regularization-inspired anisotropic term to control diffusion between nodes and incorporates a learnable parameterization for feature coupling during the diffusion process. Our model performs competitively against several GNN baselines for both heterophilous and homophilous graphs, demonstrating notable benefits for heterophilous graphs due to the learnable feature coupling.
Lsfdnet: A Single-Stage Fusion And Detection Network For Ships Using Swir And Lwir,
2025
Singapore Management University
Lsfdnet: A Single-Stage Fusion And Detection Network For Ships Using Swir And Lwir, Yanyin Guo, Runxuan An, Junwei Li, Zhiyuan Zhang
Research Collection School Of Computing and Information Systems
Traditional ship detection methods primarily rely on single-modal approaches, such as visible or infrared images, which limit their application in complex scenarios involving varying lighting conditions and heavy fog. To address this issue, we explore the advantages of short-wave infrared (SWIR) and long-wave infrared (LWIR) in ship detection and propose a novel single-stage image fusion detection algorithm called LSFDNet. This algorithm leverages feature interaction between the image fusion and object detection subtask networks, achieving remarkable detection performance and generating visually impressive fused images. To further improve the saliency of objects in the fused images and improve the performance of the …
Mitigating Cross-Modal Representation Bias For Multicultural Image-To-Recipe Retrieval,
2025
Singapore Management University
Mitigating Cross-Modal Representation Bias For Multicultural Image-To-Recipe Retrieval, Qing Wang, Chong-Wah Ngo, Yu Cao, Ee-Peng Lim
Research Collection School Of Computing and Information Systems
Existing approaches for image-to-recipe retrieval have the implicit assumption that a food image can fully capture the details textually documented in its recipe. However, a food image only reflects the visual outcome of a cooked dish and not the underlying cooking process. Consequently, learning cross-modal representations to bridge the modality gap between images and recipes tends to ignore subtle, recipe-specific details that are not visually apparent but are crucial for recipe retrieval. Specifically, the representations are biased to capture the dominant visual elements, resulting in difficulty in ranking similar recipes with subtle differences in use of ingredients and cooking methods. …
Language-Driven 3d Human Pose Estimation In Multi-Person Scenarios: A New Dataset And Approach,
2025
Singapore Management University
Language-Driven 3d Human Pose Estimation In Multi-Person Scenarios: A New Dataset And Approach, Tingrui Shen, Bangzhen Liu, Zhirun Fan, Shiting Zhang, Weifeng Pan, Sun Fan, Dan Cao, Shengfeng He
Research Collection School Of Computing and Information Systems
In an NBA game scenario, consider the challenge of locating and analyzing the 3D poses of players performing a user-specified action, such as attempting a shot. Traditional 3D human pose estimation (3DHPE) methods often fall short in such complex, multi-person scenes due to their lack of semantic integration and reliance on isolated pose data. To address these limitations, we introduce Language-Driven 3D Human Pose Estimation (L3DHPE), a novel approach that extends 3DHPE to general multi-person contexts by incorporating detailed language descriptions. We present Panoptic-L3D, the first dataset designed for L3DHPE, featuring 3,838 linguistic annotations for 1,476 individuals across 588 videos, …
Diffusionmat: Alpha Matting As Deterministic Sequential Refinement Learning,
2025
Singapore Management University
Diffusionmat: Alpha Matting As Deterministic Sequential Refinement Learning, Yangyang Xu, Shengfeng He, Wenqi Shao, Yong Du, Kwan-Yee K. Wong, Yu Qiao, Jun Yu, Ping Luo
Research Collection School Of Computing and Information Systems
In this paper, we introduce DiffusionMat, a novel image matting framework that employs a diffusion model for the transition from coarse to refined alpha mattes. Diverging from conventional methods that utilize trimaps merely as loose guidance for alpha matte prediction, our approach treats image matting as a deterministic sequential refinement learning process. This process begins with the addition of noise to trimaps and iteratively denoises them using a pre-trained diffusion model, which incrementally guides the prediction towards a clean alpha matte. The key innovation of our framework is a correction module that adjusts the output at each denoising step, ensuring …
Genwardrobe: A Fully Generative System For Travel Fashion Wardrobe Construction,
2025
Singapore Management University
Genwardrobe: A Fully Generative System For Travel Fashion Wardrobe Construction, Peng Jin, Yilin Wen, Mingzhe Yu, Yunshan Ma, Rong Zheng, Jin‑Tu Fan, Chong Wah Ngo
Research Collection School Of Computing and Information Systems
With the increasing demand for outfit planning in real-world travel scenarios, the need for constructing a travel fashion wardrobe, a series of outfits tailored to a user's personalization and destination-specific context over a short travel period, has grown significantly. However, existing systems or works often focus on isolated factors and rely on retrieval-based methods, with insufficient utilization of generative models, limiting their adaptability to real-world travel scenarios. To address this issue, this study introduces GenWardrobe, a fully generative system for travel fashion wardrobe construction. GenWardrobe consists of three key modules: user query analysis, fashion knowledge retrieval via retrieval-augmented generation and …
Look Before You Decide: Prompting Active Deduction Of Mllms For Assumptive Reasoning,
2025
Singapore Management University
Look Before You Decide: Prompting Active Deduction Of Mllms For Assumptive Reasoning, Yian Li, Wentao Tian, Yang Jiao, Jingjing Chen, Tianwen Qian, Bin Zhu, Na Zhao, Yu‑Gang Jiang
Research Collection School Of Computing and Information Systems
Recently, Multimodal Large Language Models (MLLMs) have achieved significant success across multiple disciplines due to their exceptional instruction-following capabilities and extensive world knowledge. However, whether these MLLMs possess human-like compositional reasoning abilities remains an open problem. To unveil their reasoning behaviors, we first curate a Multimodal Assumptive Reasoning Benchmark (MARS-Bench) in this paper. Interestingly, we find that most prevalent MLLMs can be easily fooled by the introduction of a presupposition into the question, whereas such presuppositions appear naive to human reasoning. Besides, we also propose a simple yet effective method, Active Deduction (AD), a novel reinforcement learning paradigm to encourage …
Cookingdiffusion: Cooking Procedural Image Generation With Stable Diffusion,
2025
Singapore Management University
Cookingdiffusion: Cooking Procedural Image Generation With Stable Diffusion, Yuan Wang, Bin Zhu, Yanbin Hao, Chong-Wah Ngo, Yi Tan, Xiang Wang
Research Collection School Of Computing and Information Systems
Recent advancements in text-to-image generation models have excelled in creating diverse and realistic images. This success extends to food imagery, where various conditional inputs like cooking styles, ingredients, and recipes are utilized. However, a yet-unexplored challenge is generating a sequence of procedural images based on cooking steps from a recipe. This could enhance the cooking experience with visual guidance and possibly lead to an intelligent cooking simulation system. To fill this gap, we introduce a novel task called cooking procedural image generation. This task is inherently demanding, as it strives to create photo-realistic images that align with cooking steps while …
From Holistic To Localized: Local Enhanced Adapters For Efficient Visual Instruction Fine-Tuning,
2025
Singapore Management University
From Holistic To Localized: Local Enhanced Adapters For Efficient Visual Instruction Fine-Tuning, Pengkun Jiao, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yugang Jiang
Research Collection School Of Computing and Information Systems
Efficient Visual Instruction Fine-Tuning (EVIT) seeks to adapt Multimodal Large Language Models (MLLMs) to downstream tasks with minimal computational overhead. However, as task diversity and complexity increase, EVIT faces significant challenges in resolving data conflicts. To address this limitation, we propose the Dual Low-Rank Adaptation (Dual-LoRA), a holistic-to-local framework that enhances the adapter’s capacity to address data conflict through dual structural optimization. Specifically, we utilize two subspaces: a skill space for stable, holistic knowledge retention, and a rank-rectified task space that locally activates the holistic knowledge. Additionally, we introduce Visual Cue Enhancement (VCE), a multi-level local feature aggregation module designed …
Exploring Object Status Recognition For Recipe Progress Tracking In Non-Visual Cooking,
2025
Singapore Management University
Exploring Object Status Recognition For Recipe Progress Tracking In Non-Visual Cooking, Franklin Mingzhe Li, Kaitlyn Ng, Bin Zhu, Patrick Carrington
Research Collection School Of Computing and Information Systems
Cooking plays a vital role in everyday independence and well-being, yet remains challenging for people with vision impairments due to limited support for tracking progress and receiving contextual feedback. Object status — the condition or transformation of ingredients and tools — offers a promising but underexplored foundation for context-aware cooking support. In this paper, we present OSCAR (Object Status Context Awareness for Recipes), a technical pipeline that explores the use of object status recognition to enable recipe progress tracking in non-visual cooking. OSCAR integrates recipe parsing, object status extraction, visual alignment with cooking steps, and time-causal modeling to support real-time …
Map As A By-Product: Collective Landmark Mapping From Imu Data And User-Provided Texts In Situated Tasks,
2025
Singapore Management University
Map As A By-Product: Collective Landmark Mapping From Imu Data And User-Provided Texts In Situated Tasks, Ryo Yonetani, Kotaro Hara
Research Collection School Of Computing and Information Systems
This paper presents Collective Landmark Mapper, a novel map-as-a-by-product system for generating semantic landmark maps of indoor environments. Consider users engaged in situated tasks that require them to navigate these environments and regularly take notes on their smartphones. Collective Landmark Mapper exploits the smartphone's IMU data and the user's free text input during these tasks to identify a set of landmarks encountered by the user. The identified landmarks are then aggregated across multiple users to generate a unified map representing the positions and semantic information of all landmarks. In developing the proposed system, we focused specifically on retail applications and …
Robface: A Test Suite For Efficient Robustness Evaluation Of Face Recognition Systems,
2025
Singapore Management University
Robface: A Test Suite For Efficient Robustness Evaluation Of Face Recognition Systems, Ruihan Zhang, Jun Sun
Research Collection School Of Computing and Information Systems
Face recognition is a widely used authentication technology in practice, where robustness is required. It is thus essential to have an efficient and easy-to-use method for evaluating the robustness of (possibly third-party) trained face recognition systems. Existing approaches to evaluating the robustness of face recognition systems are either based on empirical evaluation (e.g., measuring attacking success rate using state-of-the-art attacking methods) or formal analysis (e.g., measuring the Lipschitz constant). While the former demands significant user efforts and expertise, the latter is extremely time-consuming. In pursuit of a comprehensive, efficient, easy-to-use, and scalable estimation of the robustness of face recognition systems, …
Vibemus: Proactive Agentic System For Music Personalization,
2025
Singapore Management University
Vibemus: Proactive Agentic System For Music Personalization, Zhiliang Guo, Teng Tu, Yunshan Ma, Xun Yang
Research Collection School Of Computing and Information Systems
Large language models (LLMs) enable diverse forms of AI-assisted creation, yet they often struggle to bridge the preference-articulation gap: users may provide incomplete or vague intentions or lack the vocabulary to specify what they want, yielding outputs misaligned with true preferences. To address this gap and facilitate music creation in a vibe-centric environment, we introduce VibeMus, a proactive agentic system built on open-source components. The system engages in multi-turn dialogue to progressively determine the music’s emotion, genre, lyrics, and other aspects before generation. Simulated evaluations show that proactive clarification improves alignment with users’ intended nuances. Our approach is training-free, leveraging …
Stylegan-∞: Extending Stylegan To Arbitrary-Ratio Translation With Stylebook,
2025
South China University of Technology
Stylegan-∞: Extending Stylegan To Arbitrary-Ratio Translation With Stylebook, Yihua Dai, Tianyi Xiang, Bailin Deng, Yong Du, Hongmin Cai, Jing Qin, Shengfeng He
Research Collection School Of Computing and Information Systems
Although pre-trained large-scale generative models StyleGAN series have proven to be effective in various editing and translation tasks, they are limited to pre-defined fixed aspect ratio. To overcome this limitation, we propose StyleGAN-∞, a model that enables pre-trained StyleGAN to perform arbitrary-ratio conditional synthesis. Our key insight is to distill the expressive StyleGAN features into a StyleBook, such that an arbitrary-ratio condition can be translated to other forms by properly assembling pre-defined StyleBook vectors. To learn and leverage the StyleBook, we employ a network with three distinct stages, each corresponding to StyleBook extraction, StyleBook correspondence learning, and arbitrary-ratio synthesis. Extensive …
Mixsa: Training-Free Reference-Based Sketch Extraction Via Mixture-Of-Self-Attention,
2025
Shaanxi Normal University
Mixsa: Training-Free Reference-Based Sketch Extraction Via Mixture-Of-Self-Attention, Rui Yang, Xiaojun Wu, Shengfeng He
Research Collection School Of Computing and Information Systems
Current sketch extraction methods either require extensive training or fail to capture a wide range of artistic styles, limiting their practical applicability and versatility. We introduce Mixture-of-Self-Attention (MixSA), a training-free sketch extraction method that leverages strong diffusion priors for enhanced sketch perception. At its core, MixSA employs a mixture-of-self-attention technique, which manipulates self-attention layers by substituting the keys and values with those from reference sketches. This allows for the seamless integration of brushstroke elements into initial outline images, offering precise control over texture density and enabling interpolation between styles to create novel, unseen styles. By aligning brushstroke styles with the …
