arXiv subjects

Peiwu Qin

Publications and source records attributed to Peiwu Qin.

At least 37 records · Page 2Linked to original sources

arXiv2025

AdaDocVQA: Adaptive Framework for Long Document Visual Question Answering in Low-Resource Settings

Document Visual Question Answering (Document VQA) faces significant challenges when processing long documents in low-resource environments due to context limitations and insufficient training data. This paper presents AdaDocVQA, a unified adaptive framework addressing these challenges through three core innovations: a hybrid text retrieval architecture for effective document segmentation, an intelligent data augmentation pipeline that automatically generates high-quality reasoning question-answer pairs with multi-level verification, and adaptive ensemble inference with dynamic configuration generation and early stopping mechanisms. Experiments on Japanese document VQA benchmarks demonstrate substantial improvements with 83.04\% accuracy on Yes/No questions, 52.66\% on factual questions, and 44.12\% on numerical questions in JDocQA, and 59\% accuracy on LAVA dataset. Ablation studies confirm meaningful contributions from each component, and our framework establishes new state-of-the-art results for Japanese document VQA while providing a scalable foundation for other low-resource languages and specialized domains. Our code available at: https://github.com/Haoxuanli-Thu/AdaDocVQA.

cs.CL↗

arXiv2025

Hear-Your-Click: Interactive Object-Specific Video-to-Audio Generation

Video-to-audio (V2A) generation shows great potential in fields such as film production. Despite significant advances, current V2A methods relying on global video information struggle with complex scenes and generating audio tailored to specific objects. To address these limitations, we introduce Hear-Your-Click, an interactive V2A framework enabling users to generate sounds for specific objects by clicking on the frame. To achieve this, we propose Object-aware Contrastive Audio-Visual Fine-tuning (OCAV) with a Mask-guided Visual Encoder (MVE) to obtain object-level visual features aligned with audio. Furthermore, we tailor two data augmentation strategies, Random Video Stitching (RVS) and Mask-guided Loudness Modulation (MLM), to enhance the model's sensitivity to segmented objects. To measure audio-visual correspondence, we designed a new evaluation metric, the CAV score. Extensive experiments demonstrate that our framework offers more precise control and improves generation performance across various metrics. Project Page: https://github.com/SynapGrid/Hear-Your-Click

cs.CV↗

arXiv2025

pLDDT-Predictor: High-speed Protein Screening Using Transformer and ESM2

Recent advancements in protein structure prediction, particularly AlphaFold2, have revolutionized structural biology by achieving near-experimental accuracy ($\text{average RMSD} < 1.5\textÅ$). However, the computational demands of these models (approximately 30 minutes per protein on an RTX 4090) significantly limit their application in high-throughput protein screening. While large language models like ESM (Evolutionary Scale Modeling) have shown promise in extracting structural information directly from protein sequences, rapid assessment of protein structure quality for large-scale analyses remains a major challenge. We introduce pLDDT-Predictor, a high-speed protein screening tool that achieves a $250,000\times$ speedup compared to AlphaFold2 by leveraging pre-trained ESM2 protein embeddings and a Transformer architecture. Our model predicts AlphaFold2's pLDDT (predicted Local Distance Difference Test) scores with a Pearson correlation of 0.7891 and processes proteins in just 0.007 seconds on average. Using a comprehensive dataset of 1.5 million diverse protein sequences (ranging from 50 to 2048 amino acids), we demonstrate that pLDDT-Predictor accurately classifies high-confidence structures (pLDDT $>$ 70) with 91.2\% accuracy and achieves an MSE of 84.8142 compared to AlphaFold2's predictions. The source code and pre-trained models are freely available at https://github.com/jw-chae/pLDDT_Predictor, enabling the research community to perform rapid, large-scale protein structure quality assessments.

q-bio.BM↗

arXiv2025

Dual-Modality Computational Ophthalmic Imaging with Deep Learning and Coaxial Optical Design

The growing burden of myopia and retinal diseases necessitates more accessible and efficient eye screening solutions. This study presents a compact, dual-function optical device that integrates fundus photography and refractive error detection into a unified platform. The system features a coaxial optical design using dichroic mirrors to separate wavelength-dependent imaging paths, enabling simultaneous alignment of fundus and refraction modules. A Dense-U-Net-based algorithm with customized loss functions is employed for accurate pupil segmentation, facilitating automated alignment and focusing. Experimental evaluations demonstrate the system's capability to achieve high-precision pupil localization (EDE = 2.8 px, mIoU = 0.931) and reliable refractive estimation with a mean absolute error below 5%. Despite limitations due to commercial lens components, the proposed framework offers a promising solution for rapid, intelligent, and scalable ophthalmic screening, particularly suitable for community health settings.

eess.IV↗

arXiv2025

HCMA-UNet: A Hybrid CNN-Mamba UNet with Axial Self-Attention for Efficient Breast Cancer Segmentation

Breast cancer lesion segmentation in DCE-MRI remains challenging due to heterogeneous tumor morphology and indistinct boundaries. To address these challenges, this study proposes a novel hybrid segmentation network, HCMA-UNet, for lesion segmentation of breast cancer. Our network consists of a lightweight CNN backbone and a Multi-view Axial Self-Attention Mamba (MISM) module. The MISM module integrates Visual State Space Block (VSSB) and Axial Self-Attention (ASA) mechanism, effectively reducing parameters through Asymmetric Split Channel (ASC) strategy to achieve efficient tri-directional feature extraction. Our lightweight model achieves superior performance with 2.87M parameters and 126.44 GFLOPs. A Feature-guided Region-aware loss function (FRLoss) is proposed to enhance segmentation accuracy. Extensive experiments on one private and two public DCE-MRI breast cancer datasets demonstrate that our approach achieves state-of-the-art performance while maintaining computational efficiency. FRLoss also exhibits good cross-architecture generalization capabilities. The source code is available at https://github.com/Haoxuanli-Thu/HCMA-UNet.

eess.IV↗

arXiv2025

DBF-UNet: A Two-Stage Framework for Carotid Artery Segmentation with Pseudo-Label Generation

Medical image analysis faces significant challenges due to limited annotation data, particularly in three-dimensional carotid artery segmentation tasks, where existing datasets exhibit spatially discontinuous slice annotations with only a small portion of expert-labeled slices in complete 3D volumetric data. To address this challenge, we propose a two-stage segmentation framework. First, we construct continuous vessel centerlines by interpolating between annotated slice centroids and propagate labels along these centerlines to generate interpolated annotations for unlabeled slices. The slices with expert annotations are used for fine-tuning SAM-Med2D, while the interpolated labels on unlabeled slices serve as prompts to guide segmentation during inference. In the second stage, we propose a novel Dense Bidirectional Feature Fusion UNet (DBF-UNet). This lightweight architecture achieves precise segmentation of complete 3D vascular structures. The network incorporates bidirectional feature fusion in the encoder and integrates multi-scale feature aggregation with dense connectivity for effective feature reuse. Experimental validation on public datasets demonstrates that our proposed method effectively addresses the sparse annotation challenge in carotid artery segmentation while achieving superior performance compared to existing approaches. The source code is available at https://github.com/Haoxuanli-Thu/DBF-UNet.

cs.CV↗

arXiv2024

Grid-augmented vision: A simple yet effective approach for enhanced spatial understanding in multi-modal agents

Recent advances in multimodal models have demonstrated impressive capabilities in object recognition and scene understanding. However, these models often struggle with precise spatial localization - a critical capability for real-world applications. Inspired by how humans use grid-based references like chess boards and maps, we propose introducing explicit visual position encoding through a simple grid overlay approach. By adding a 9x9 black grid pattern onto input images, our method provides visual spatial guidance analogous to how positional encoding works in transformers, but in an explicit, visual form. Experiments on the COCO 2017 dataset demonstrate that our grid-based approach achieves significant improvements in localization accuracy, with a 107.4% increase in IoU (from 0.27 to 0.56) and a 194.4% improvement in GIoU (from 0.18 to 0.53) compared to baseline performance. Through attention visualization analysis, we show how this visual position encoding helps models better ground spatial relationships. Our method's simplicity and effectiveness make it particularly valuable for applications requiring accurate spatial reasoning, such as robotic manipulation, medical imaging, and autonomous navigation.

cs.CV↗

arXiv2024

Practical Guidelines for Cell Segmentation Models Under Optical Aberrations in Microscopy

Cell segmentation is essential in biomedical research for analyzing cellular morphology and behavior. Deep learning methods, particularly convolutional neural networks (CNNs), have revolutionized cell segmentation by extracting intricate features from images. However, the robustness of these methods under microscope optical aberrations remains a critical challenge. This study evaluates cell image segmentation models under optical aberrations from fluorescence and bright field microscopy. By simulating different types of aberrations, including astigmatism, coma, spherical aberration, trefoil, and mixed aberrations, we conduct a thorough evaluation of various cell instance segmentation models using the DynamicNuclearNet (DNN) and LIVECell datasets, representing fluorescence and bright field microscopy cell datasets, respectively. We train and test several segmentation models, including the Otsu threshold method and Mask R-CNN with different network heads (FPN, C3) and backbones (ResNet, VGG, Swin Transformer), under aberrated conditions. Additionally, we provide usage recommendations for the Cellpose 2.0 Toolbox on complex cell degradation images. The results indicate that the combination of FPN and SwinS demonstrates superior robustness in handling simple cell images affected by minor aberrations. In contrast, Cellpose 2.0 proves effective for complex cell images under similar conditions. Furthermore, we innovatively propose the Point Spread Function Image Label Classification Model (PLCM). This model can quickly and accurately identify aberration types and amplitudes from PSF images, assisting researchers without optical training. Through PLCM, researchers can better apply our proposed cell segmentation guidelines.

eess.IV↗

arXiv2024

External Prompt Features Enhanced Parameter-efficient Fine-tuning for Salient Object Detection

Salient object detection (SOD) aims at finding the most salient objects in images and outputs pixel-level binary masks. Transformer-based methods achieve promising performance due to their global semantic understanding, crucial for identifying salient objects. However, these models tend to be large and require numerous training parameters. To better harness the potential of transformers for SOD, we propose a novel parameter-efficient fine-tuning method aimed at reducing the number of training parameters while enhancing the salient object detection capability. Our model, termed EXternal Prompt features Enhanced adapteR Tuning (ExPert), features an encoder-decoder structure with adapters and injectors interspersed between the layers of a frozen transformer encoder. The adapter modules adapt the pretrained backbone to SOD while the injector modules incorporate external prompt features to enhance the awareness of salient objects. Comprehensive experiments demonstrate the superiority of our method. Surpassing former state-of-the-art (SOTA) models across five SOD datasets, ExPert achieves 0.215 mean absolute error (MAE) in the ECSSD dataset with 80.2M trained parameters, 21% better than SelfReformer and 47% better than EGNet.

cs.CV↗

arXiv2024

Harnessing Intra-group Variations Via a Population-Level Context for Pathology Detection

Realizing sufficient separability between the distributions of healthy and pathological samples is a critical obstacle for pathology detection convolutional models. Moreover, these models exhibit a bias for contrast-based images, with diminished performance on texture-based medical images. This study introduces the notion of a population-level context for pathology detection and employs a graph theoretic approach to model and incorporate it into the latent code of an autoencoder via a refinement module we term PopuSense. PopuSense seeks to capture additional intra-group variations inherent in biomedical data that a local or global context of the convolutional model might miss or smooth out. Proof-of-concept experiments on contrast-based and texture-based images, with minimal adaptation, encounter the existing preference for intensity-based input. Nevertheless, PopuSense demonstrates improved separability in contrast-based images, presenting an additional avenue for refining representations learned by a model.

eess.IV↗

arXiv2024

Deep learning enables automated assessments of inflammatory response in zebrafish exposed to different pollutants

In the field of environmental toxicology, rapid and precise assessment of the inflammatory response to pollutants in biological models is critical. This study leverages the power of deep learning to enable automated assessments of zebrafish, a model organism widely used for its translational relevance to human disease pathways. We present an innovative approach to assessing inflammatory responses in zebrafish exposed to various pollutants through an end-to-end deep learning model. The model employs a Unet-based architecture to automatically process high-throughput lateral zebrafish images, segmenting specific regions and quantifying neutrophils as inflammation markers. Alongside imaging, qPCR analysis offers gene expression insights, revealing the molecular impact of exposure on inflammatory pathways. Moreover, the deep learning model was packaged as a user-friendly executable file (.exe), facilitating widespread application by enabling use on virtually any computer without the need for specialized software or training.

physics.bio-ph↗

arXiv2024

COVID-19: post infection implications in different age groups, mechanism, diagnosis, effective prevention, treatment, and recommendations

SARS-CoV-2, the highly contagious pathogen responsible for the COVID-19 pandemic, has persistent effects that begin four weeks after initial infection and last for an undetermined duration. These chronic effects are more harmful than acute ones. This review explores the long-term impact of the virus on various human organs, including the pulmonary, cardiovascular, neurological, reproductive, gastrointestinal, musculoskeletal, endocrine, and lymphoid systems, particularly in older adults. Regarding diagnosis, RT-PCR is the gold standard for detecting COVID-19, though it requires specialized equipment, skilled personnel, and considerable time to produce results. To address these limitations, artificial intelligence in imaging and microfluidics technologies offers promising alternatives for diagnosing COVID-19 efficiently. Pharmacological and non-pharmacological strategies are effective in mitigating the persistent impacts of COVID-19. These strategies enhance immunity in post-COVID-19 patients by reducing cytokine release syndrome, improving T cell response, and increasing the circulation of activated natural killer and CD8 T cells in blood and tissues. This, in turn, alleviates symptoms such as fever, nausea, fatigue, muscle weakness, and pain. Vaccines, including inactivated viral, live attenuated viral, protein subunit, viral vectored, mRNA, DNA, and nanoparticle vaccines, significantly reduce the adverse long-term effects of the virus. However, no vaccine has been reported to provide lifetime protection against COVID-19. Consequently, protective measures such as physical distancing, mask usage, and hand hygiene remain essential strategies. This review offers a comprehensive understanding of the persistent effects of COVID-19 on individuals of varying ages, along with insights into diagnosis, treatment, vaccination, and future preventative measures against the spread of SARS-CoV-2.

q-bio.QM↗

arXiv2024

Application and prospect of hydrogels in diabetic wound treatment

Diabetic foot patients face persistent, challenging wounds, prompting a quest for innovative therapies. Hydrogel, a promising modality, is increasingly recognized for its potential in diabetic wound care. This review examines hydrogel's emergence as a therapeutic option, and its benefits, and outlines research directions. Diabetic foot ulcers, notoriously hard to heal, demand effective solutions beyond traditional methods. Hydrogel, with its hydrophilic nature and biocompatibility, offers a novel approach. It fosters a moist wound environment, aids healing, and mitigates infection risks. Additionally, its versatility in delivering bioactive agents enhances wound recovery. This analysis provides insights into hydrogel's application, paving the way for improved outcomes and patient well-being in diabetic foot care.

physics.med-ph↗

arXiv2024

Pneumonia App: a mobile application for efficient pediatric pneumonia diagnosis using explainable convolutional neural networks (CNN)

Mycoplasma pneumoniae pneumonia (MPP) poses significant diagnostic challenges in pediatric healthcare, especially in regions like China where it's prevalent. We introduce PneumoniaAPP, a mobile application leveraging deep learning techniques for rapid MPP detection. Our approach capitalizes on convolutional neural networks (CNNs) trained on a comprehensive dataset comprising 3345 chest X-ray (CXR) images, which includes 833 CXR images revealing MPP and additionally augmented with samples from a public dataset. The CNN model achieved an accuracy of 88.20% and an AUROC of 0.9218 across all classes, with a specific accuracy of 97.64% for the mycoplasma class, as demonstrated on the testing dataset. Furthermore, we integrated explainability techniques into PneumoniaAPP to aid respiratory physicians in lung opacity localization. Our contribution extends beyond existing research by targeting pediatric MPP, emphasizing the age group of 0-12 years, and prioritizing deployment on mobile devices. This work signifies a significant advancement in pediatric pneumonia diagnosis, offering a reliable and accessible tool to alleviate diagnostic burdens in healthcare settings.

eess.IV↗

arXiv2024

LAMPER: LanguAge Model and Prompt EngineeRing for zero-shot time series classification

This study constructs the LanguAge Model with Prompt EngineeRing (LAMPER) framework, designed to systematically evaluate the adaptability of pre-trained language models (PLMs) in accommodating diverse prompts and their integration in zero-shot time series (TS) classification. We deploy LAMPER in experimental assessments using 128 univariate TS datasets sourced from the UCR archive. Our findings indicate that the feature representation capacity of LAMPER is influenced by the maximum input token threshold imposed by PLMs.

cs.AI↗

arXiv2024

Cognitive resilience: Unraveling the proficiency of image-captioning models to interpret masked visual content

This study explores the ability of Image Captioning (IC) models to decode masked visual content sourced from diverse datasets. Our findings reveal the IC model's capability to generate captions from masked images, closely resembling the original content. Notably, even in the presence of masks, the model adeptly crafts descriptive textual information that goes beyond what is observable in the original image-generated captions. While the decoding performance of the IC model experiences a decline with an increase in the masked region's area, the model still performs well when important regions of the image are not masked at high coverage.

cs.CV↗

arXiv2024

IRFundusSet: An Integrated Retinal Fundus Dataset with a Harmonized Healthy Label

Ocular conditions are a global concern and computational tools utilizing retinal fundus color photographs can aid in routine screening and management. Obtaining comprehensive and sufficiently sized datasets, however, is non-trivial for the intricate retinal fundus, which exhibits heterogeneities within pathologies, in addition to variations from demographics and acquisition. Moreover, retinal fundus datasets in the public space suffer fragmentation in the organization of data and definition of a healthy observation. We present Integrated Retinal Fundus Set (IRFundusSet), a dataset that consolidates, harmonizes and curates several public datasets, facilitating their consumption as a unified whole and with a consistent is_normal label. IRFundusSet comprises a Python package that automates harmonization and avails a dataset object in line with the PyTorch approach. Moreover, images are physically reviewed and a new is_normal label is annotated for a consistent definition of a healthy observation. Ten public datasets are initially considered with a total of 46064 images, of which 25406 are curated for a new is_normal label and 3515 are deemed healthy across the sources.

cs.CV↗

arXiv2023

Light sheet and light field microscopy based on scanning Bessel beam illumination

We developed a Bessel light sheet fluorescence microscopy (LSFM) system to enable high-speed, wide-field intra-vital imaging of zebrafish and other thick biological samples. This system uses air objectives for the convenient mounting of large samples and incorporates an electrically tunable lens for automatic focusing during volumetric imaging. To enhance the precision of 3D imaging, the impact of the electrically tunable lens on system magnification is investigated and modified through designed experiments. Despite using Bessel beams with side lobes, we achieved satisfactory image quality through a straightforward background noise subtraction method, eliminating the need for further deconvolution. Our system provides zebrafish imaging at a resolution comparable to commercial confocal microscopy but in just 1/40th of the time. We also introduced light field microscopy (LFM) to improve 3D in vivo imaging temporal resolution. Apart from the 28-fold speed enhancement, the comparison of LFM and LSFM results reveals a unique aspect of LFM imaging concerning image dynamic range, which has not been previously reported.

physics.optics↗