Search Results (521)

Urban water bodies are crucial in urban planning and flood detection, and they are susceptible to changes due to climate change and rapid urbanization. With the development of high-resolution remote sensing technology and the success of semantic segmentation using deep learning in computer vision, it is possible to extract urban water bodies from high-resolution remote sensing images. However, many urban water bodies are small, oddly shaped, silted, or spectrally similar to other objects, making their extraction extremely challenging. In this paper, we propose a neural network named CM-UNet++, a combination of the dense-skip module based on UNet++ and the CSMamba module to encode different levels’ information with interactions and then extract global and local information at each level. We use a size-weighted auxiliary loss function to balance feature maps of different levels. Additionally, features beyond RGB are incorporated into the input of the neural network to enhance the distinction between water bodies and other objects. We produced a labeled urban water extraction dataset, and experiments on this dataset show that CM-UNet++ attains 0.8781 on the IOU (intersection over union) metric, which indicates that this method outperforms other recent semantic segmentation methods and achieves better completeness, connectivity, and boundary accuracy. The proposed dense-skip module and CSMamba module significantly improve the extraction of small and spectrally indistinct water bodies. Furthermore, experiments on a public dataset confirm the method’s robustness. Full article

Hand gesture recognition (HGR) is a convenient and natural form of human–computer interaction. It is suitable for various applications. Much research has already focused on wearable device-based HGR. By contrast, this paper gives an overview focused on device-free HGR. That means we evaluate HGR systems that do not require the user to wear something like a data glove or hold a device. HGR systems are explored regarding technology, hardware, and algorithms. The interconnectedness of timing and power requirements with hardware, pre-processing algorithm, classification, and technology and how they permit more or less granularity, accuracy, and number of gestures is clearly demonstrated. Sensor modalities evaluated are WIFI, vision, radar, mobile networks, and ultrasound. The pre-processing technologies stereo vision, multiple-input multiple-output (MIMO), spectrogram, phased array, range-doppler-map, range-angle-map, doppler-angle-map, and multilateration are explored. Classification approaches with and without ML are studied. Among those with ML, assessed algorithms range from simple tree structures to transformers. All applications are evaluated taking into account their level of integration. This encompasses determining whether the application presented is suitable for edge integration, their real-time capability, whether continuous learning is implemented, which robustness was achieved, whether ML is applied, and the accuracy level. Our survey aims to provide a thorough understanding of the current state of the art in device-free HGR on edge devices and in general. Finally, on the basis of present-day challenges and opportunities in this field, we outline which further research we suggest for HGR improvement. Our goal is to promote the development of efficient and accurate gesture recognition systems. Full article

Recognizing human activities from motion data is a complex task in computer vision, involving the recognition of human behaviors from sequences of 3D motion data. These activities encompass successive body part movements, interactions with objects, or group dynamics. Camera-based recognition methods are cost-effective and perform well under controlled conditions but face challenges in real-world scenarios due to factors such as viewpoint changes, illumination variations, and occlusion. The latter is the most significant challenge in real-world recognition; partial occlusion impacts recognition accuracy to varying degrees depending on the activity and the occluded body parts while complete occlusion can render activity recognition impossible. In this paper, we propose a novel approach for human activity recognition in the presence of partial occlusion, which may be applied in cases wherein up to two body parts are occluded. The proposed approach works under the assumptions that (a) human motion is modeled using a set of 3D skeletal joints, and (b) the same body parts remain occluded throughout the whole activity. Contrary to previous research, in this work, we address this problem using a Generative Adversarial Network (GAN). Specifically, we train a Convolutional Recurrent Neural Network (CRNN), whose goal is to serve as the generator of the GAN. Its aim is to complete the missing parts of the skeleton due to occlusion. Specifically, the input to this CRNN consists of raw 3D skeleton joint positions, upon the removal of joints corresponding to occluded parts. The output of the CRNN is a reconstructed skeleton. For the discriminator of the GAN, we use a simple long short-term memory (LSTM) network. We evaluate the proposed approach using publicly available datasets in a series of occlusion scenarios. We demonstrate that in all scenarios, the occlusion of certain body parts causes a significant decline in performance, although in some cases, the reconstruction process leads to almost perfect recognition. Nonetheless, in almost every circumstance, the herein proposed approach exhibits superior performance compared to previous works, which varies between $2.2\%$ and $37.5\%$, depending on the dataset used and the occlusion case. Full article

The solution to the problem of insufficient accuracy in determining the position and speed of human movement during interaction with a treadmill-based training complex is considered. Control command generation based on the training complex user’s actions may be performed with a delay, may not take into account the specificity of movements, or be inaccurate due to the error of the initial data. The article introduces a technology for improving the accuracy of predicting a person’s position and speed on a running platform using machine learning and computer vision methods. The proposed technology includes analysing and processing data from the tracking system, developing machine learning models to improve the quality of the raw data, predicting the position and speed of human movement, and implementing and integrating neural network methods into the running platform control system. Experimental results demonstrate that the decision tree (DT) model provides better accuracy and performance in solving the problem of positioning key points of a human model in complex conditions with overlapping limbs. For speed prediction, the linear regression (LR) model showed the best results when the analysed window length was 10 frames. Prediction of the person’s position (based on 10 previous frames) is performed using the DT model, which is optimal in terms of accuracy and computation time relative to other options. The comparison of the control methods of the running platform based on machine learning models showed the advantage of the combined method (linear control function combined with the speed prediction model), which provides an average absolute error value of 0.116 m/s. The results of the research confirmed the achievement of the primary objective (increasing the accuracy of human position and speed prediction), making the proposed technology promising for application in human-machine systems. Full article

Human communication has long relied on visual media for interaction, and is facilitated by electronic devices that access visual data. Traditionally, this exchange was unidirectional, constrained to text-based queries. However, advancements in human–computer interaction have introduced technologies like reverse image search and large language models (LLMs), enabling both textual and visual queries. These innovations are particularly valuable in Cultural Heritage applications, such as connecting tourists with point-of-interest recognition systems during city visits. This paper investigates the use of various Vision Language Models (VLMs) for Cultural Heritage visual question aswering, including Bing’s search engine with GPT-4 and open models such as Qwen2-VL and Pixtral. Twenty Italian landmarks were selected for the study, including the Colosseum, Milan Cathedral, and Michelangelo’s David. For each landmark, two images were chosen: one from Wikipedia and another from a scientific database or private collection. These images were input into each VLM with textual queries regarding their content. We studied the quality of the responses in terms of their completeness, assessing the impact of various levels of detail in the queries. Additionally, we explored the effect of language (English vs. Italian) on the models’ ability to provide accurate answers. Our findings indicate that larger models, such as Qwen2-VL and Bing+ChatGPT-4, which are trained on multilingual datasets, perform better in both English and Italian. Iconic landmarks like the Colosseum and Florence’s Duomo are easily recognized, and providing context (e.g., the city) improves identification accuracy. Surprisingly, the Wikimedia dataset did not perform as expected, with varying results across models. Open models like Qwen2-VL, which can run on consumer workstations, showed performance similar to larger models. While the algorithms demonstrated strong results, they also generated occasional hallucinated responses, highlighting the need for ongoing refinement of AI systems for Cultural Heritage applications. Full article

Dynamic facial expression recognition (DFER) is one of the most important challenges in computer vision, as it plays a crucial role in human–computer interaction. Recently, adapter-based approaches have been introduced into DFER, and they have achieved remarkable success. However, the adapters still suffer from the following problems: overlooking irrelevant frames and interference with pre-trained information. In this paper, we propose a frame recalibration unit adapter (FRU-Adapter) which combines the strengths of a frame recalibration unit (FRU) and temporal self-attention (T-SA) to address the aforementioned issues. The FRU initially recalibrates the frames by emphasizing important frames and suppressing less relevant frames. The recalibrated frames are then fed into T-SA to capture the correlations between meaningful frames. As a result, the FRU-Adapter captures enhanced temporal dependencies by considering the irrelevant frames in a clip. Furthermore, we propose a method for attaching the FRU-Adapter to each encoder layer in parallel to reduce the loss of pre-trained information. Notably, the FRU-Adapter uses only 2% of the total training parameters per task while achieving an improved accuracy. Extended experiments on DFER tasks show that the proposed FRU-Adapter not only outperforms the state-of-the-art models but also exhibits parameter efficiency. The source code will be made publicly available. Full article

Background/Objectives: RNA research is critical for understanding gene regulation, disease mechanisms, and therapeutic development. Constructing effective RNA benchmark models for accurate downstream analysis has become a significant research challenge. The objective of this study is to propose a robust benchmark model, DRFormer, for RNA sequence downstream tasks. Methods: The DRFormer model utilizes RNA sequences to construct novel vision features based on secondary structure and sequence distance. These features are pre-trained using the SWIN model to develop a SWIN-RNA submodel. This submodel is then integrated with an RNA sequence model to construct a multimodal model for downstream analysis. Results: We conducted experiments on various RNA downstream tasks. In the sequence classification task, the MCC reached 94.4%, surpassing the state-of-the-art RNAErnie model by 1.2%. In the protein–RNA interaction prediction, DRFormer achieved an MCC of 0.492, outperforming advanced models like BERT-RBP and PrismNet. In RNA secondary structure prediction, the F1 score was 0.690, exceeding the widely used SPOT-RNA model by 1%. Additionally, generalization experiments on DNA tasks yielded satisfactory results. Conclusions: DRFormer is the first RNA sequence downstream analysis model that leverages structural features to construct a vision model and integrates sequence and vision models in a multimodal manner. This approach yields excellent prediction and analysis results, making it a valuable contribution to RNA research. Full article

The association between 17β-estradiol (E2) deprivation, seen in menopause, and a risk for developing glaucoma has been shown. Thus, exogenous supplementation of E2 may protect against retinal ganglion cell (RGC) degradation and vision loss. Here, we investigated the utility of topical 10β,17β-dihydroxyestra-1,4-dien-3-one (DHED), a prodrug of E2 that selectively produces the neuroprotective hormone in the retina, on visual function after optic nerve crush (ONC) and ovariectomy (OVX). We used female Brown Norway rats that underwent either Sham or OVX surgeries. After ONC, OVX animals received DHED or vehicle eye drops for 12 weeks. Visual function, via the optomotor reflex, and retinal thickness, via optical coherence tomography, were followed longitudinally. Afterward, we performed mass spectrometry-based label-free retina proteomics to survey retinal protein interaction networks in our selected animal model and to identify E2-responsive proteins after OVX on neurodegeneration. We found that ONC with OVX caused a significant decline in visual functions that were ameliorated by DHED treatments. Discovery-driven retina proteomics identified numerous proteins associated with neurodegenerative processes due to ONC that were remediated by DHED eye drops. Altogether, our three-pronged phenotypic preclinical evaluation of the topical DHED in the OVX + ONC model of glaucoma reveals the therapeutic potential of the prodrug to prevent visual deficits after glaucomatous retinal injury. Full article

Video portrait segmentation is essential for intelligent sensing systems, including human-computer interaction, autonomous navigation, and augmented reality. However, dynamic video environments introduce significant challenges, such as temporal variations, occlusions, and computational constraints. This study introduces RVM+, an enhanced video segmentation framework based on the Robust Video Matting (RVM) architecture. By incorporating Convolutional Gated Recurrent Units (ConvGRU), RVM+ improves temporal consistency and captures intricate temporal dynamics across video frames. Additionally, a novel knowledge distillation strategy reduces computational demands while maintaining high segmentation accuracy, making the framework ideal for real-time applications in resource-constrained environments. Comprehensive evaluations on challenging datasets show that RVM+ outperforms state-of-the-art methods in both segmentation accuracy and temporal consistency. Key performance indicators such as MIoU, SAD, and dtSSD effectively verify the robustness and efficiency of the model. The integration of knowledge distillation ensures a streamlined and effective design with negligible accuracy trade-offs, highlighting its suitability for practical deployment. This study makes significant strides in intelligent sensor technology, providing a high-performance, efficient, and scalable solution for video segmentation. RVM+ offers potential for applications in fields such as augmented reality, robotics, and real-time video analysis, while also advancing the development of AI-enabled vision sensors. Full article

Laparoscopic cholecystectomy (LC) is the standard procedure for gallbladder removal, but improper identification of anatomical structures can lead to biliary duct injury (BDI). The critical view of safety (CVS) is a standardized technique designed to mitigate this risk. However, existing surgical training systems primarily emphasize haptic feedback and physical skill development, making them expensive and less accessible. This paper presents the next-generation Portable Camera-Aided Surgical Simulator (PortCAS), a cost-effective, portable, vision-based surgical training simulator designed to enhance cognitive skill acquisition in LC. The system consists of an enclosed physical module equipped with a vision system, a single-board computer for real-time instrument tracking, and a virtual simulation interface that runs on a user-provided computer. Unlike traditional simulators, PortCAS prioritizes cognitive training over force-based interactions, eliminating the need for costly haptic components. The system was evaluated through user studies assessing accuracy, usability, and training effectiveness. Results demonstrate that PortCAS provides a sufficiently accurate tracking performance for training surgical skills such as CVS, offering a scalable and accessible solution for surgical education. Full article

The rapid growth of the human population, the increase in consumer needs regarding food authenticity, and the sub-par synchronization between agricultural and food industry production necessitate the development of reliable track and tracing solutions for food commodities. The present research proposes a simple and affordable digital system that could be implemented in most production processes to improve transparency and productivity. The system combines non-destructive, rapid quality assessment methods, such as near infrared spectroscopy (NIRS) and computer/machine vision (CV/MV), with track and tracing functionalities revolving around the Internet of Things (IoT) and radio frequency identification (RFID). Meanwhile, authenticity is provided by a self-developed blockchain-based solution that validates all data and documentation “from farm to fork”. The system is introduced by taking certified Hungarian sweet potato production as a model scenario. Each element of the proposed system is discussed in detail individually and as a part of an integrated system, capable of automatizing most production flows while maintaining complete transparency and compliance with authority requirements. The results include the data and trust model of the system with sequence diagrams simulating the interactions between participants. The study lays the groundwork for future research and industrial applications combining digital tools to improve the productivity and authenticity of the agri-food industry, potentially increasing the level of trust between participants, most importantly for the consumers. Full article

After criticizing the three traditional proofs of divine existence in the first Critique, Kant fills this void with an apologetic argument based on his practical philosophy. However, this moral–religious project has long been charged with various inconsistencies, particularly regarding the tension between the demand for moral perfection and human limitation. There is even some indication that he becomes aware of these issues, as he later moves away from the vision of endless moral progress that holds his original project together. However, this revision does not resolve all the tensions, as the question of how imperfect humans can be well-pleasing to God remains. I argue that this predicament is a difficult-to-avoid feature of his project given how it interacts with his religious context of Lutheran Christianity. This is because he incorporates some of its elements (particularly its uncompromising moral standard) virtually intact while radically altering others (such as vicarious atonement and imputation of alien righteousness). However, this procedure undermines the coherence of the tradition he inherits because the elements he fully incorporates are meant to lead to the traditional doctrines he leaves behind. I conclude by reflecting on how theists who are sympathetic to Kant should lead his moral–religious project out of its current precarious predicament. Full article

The purpose of this study is to investigate the influence of environmental disclosure and corporate governance on the financial performance of Islamic banks in Saudi Arabia. This study highlights that sustainable practices are transparent with financial objectives using the religious framework of Islamic finance. This research is based on Worldwide Vision 2030, which covers sustainable development and promotes environmental, social, and governance (ESG) principles, as well as corporate governance factors, such as board composition and Shariah Supervisory Boards (SSBs). We use a hybrid approach for our findings, with a dataset spanning 2011–2023 for the quantitative analysis and 20 semi-structured analyses conducted for a qualitative approach that aligns with objectives. We found that environmental disclosure boosts profits and stakeholder trust. Corporate governance structures, such as environmental boards and sustainability committees, improve the environmental disclosure of financial performance in Islamic banks. In this positive interaction, specialized governance drives Sharia-compliant sustainability initiatives. SSBs help Islamic banks integrate sustainability and meet religious and ESG environmental standards. Board diversity and dedication in the sustainability committee both play important roles in enhancing environmental disclosure practices; in return, these improved financial performances. The interaction of environmental disclosure and board environmental expertise has a positive impact on the overall performance, which indicates that governance structure supports sustainability-related decision-making, aligning with transparency. This study suggests that Islamic banks standardize ESG frameworks, improve board environmental expertise, and invest in real-time sustainability reporting digital solutions. Saudi Islamic banks can lead regional and global sustainable banking by adopting these strategies to align with global sustainability trends, improve financial performance, and meet ethical finance expectations. Full article

Recently, prompt learning has emerged as a viable technique for fine-tuning pre-trained vision–language models (VLMs). The use of prompts allows pre-trained VLMs to be quickly adapted to specific downstream tasks, bypassing the necessity to update the original pre-trained weights. Nevertheless, much of the existing work on prompt learning has focused primarily on the utilization of non-specific prompts, with little attention paid to the category-specific data. In this paper, we present a novel method, the Category-Specific Prompt (CSP), which integrates task-oriented information into our model, thereby augmenting its capacity to comprehend and execute complex tasks. In order to enhance the exploitation of features, thereby optimizing the utilization of the combination of category-specific and non-specific prompts, we introduce a novel deep prompt-learning method, Deep Contextual Prompt Enhancement (DCPE). DCPE outputs features with rich text embedding knowledge that changes in response to input through attention-based interactions, thereby ensuring that our model contains instance-oriented information. Combining the above two methods, our architecture CSP-DCPE contains both task-oriented and instance-oriented information, and achieves state-of-the-art average scores on 11 benchmark image-classification datasets. Full article

This study aims to investigate the potential impact of commercial low-vision filters on intrinsically photosensitive retinal ganglion cells (ipRGCs), which have significantly advanced our understanding of non-image-forming visual functions. A comprehensive analysis by modeling the potential responses of ipRGCs to commercially available low-vision filters was conducted, focusing on how the spectral properties of these filters could alter ipRGC function. Additionally, the influence of aging on the crystalline lens was considered. Colorimetric changes in the transmitted light by these filters were also analyzed, highlighting variations based on the manufacturer. The study uncovered the diverse responses of ipRGCs to fifty low-vision filters, shedding light on the potential modifications in ipRGC stimulation and visual function. Notably, the consideration of aging in the crystalline lens revealed significant alterations in ipRGC response. Furthermore, the analysis of colorimetric changes demonstrated substantial differences in the light transmitted by these filters, with variations dependent on the manufacturer. This research underscores the nuanced relationship between low-vision filters and ipRGCs, providing insights into their potential impact on visual function. The varying responses observed, coupled with the influence of aging on the crystalline lens, emphasize the complexity of this interaction. Additionally, the distinct colorimetric changes based on filter manufacturer suggest the need for tailored approaches in enhancing visual perception for individuals with visual impairments. Full article