Computer Vision Research Articles

Latent diffusion model for conditional reservoir facies generation

Creating accurate and geologically realistic reservoir facies based on limited measurements is crucial for field development and reservoir management, especially in the oil and gas sector. Traditional two-point geostatistics, while foundational, often struggle to capture complex geological patterns. Multi-point statistics offers more flexibility, but comes with its own challenges related to pattern configurations and storage limits. With the rise of Generative Adversarial Networks (GANs) and their success in various fields, there has been a shift towards using them for facies generation. However, recent advances in the computer vision domain have shown the superiority of diffusion models over GANs. Motivated by this, a novel Latent Diffusion Model is proposed, which is specifically designed for conditional generation of reservoir facies. The proposed model produces high-fidelity facies realizations that rigorously preserve conditioning data. It significantly outperforms a GAN-based alternative. Our implementation on GitHub: https://released-on-accept.

Study on Measurement Method of Three-dimensional Position of Unlabeled Microspheres under Bright Background

Abstract: Using computer vision technology to obtain the position and trajectory data of particle probe microspheres from microscope images has significance and value in the molecular field. However, most of the existing microsphere measurement methods are based on transmission, which can only be measured under transparent samples and substrates and are not suitable for the application scenario of living cell measurement. In this paper, a method based on reflectivity imaging is proposed to measure the three-dimensional position of the dark microspheres in the bright field. Based on the outermost ring radius method, the relationship between the inner ring radius of the microsphere spot and the out-of-focus distance was explored to measure the coordinates in the Z direction. Cardiomyocytes were combined with 10um size silica microspheres. Experiments show that in a bright field with a high perturbation environment, it can achieve high precision measurement of dark microspheres and achieve three-dimensional position measurement with an accuracy of 50nm in XY direction and 100nm in Z direction.

EXPERIMENTAL ANALYSIS OF POISSON RATIO ON METAMATERIALS MANUFACTURED FROM CRUMPLED ALUMINUM FOIL SUBJECTED TO UNIAXIAL COMPRESSION

In this work experimental results are presented concerning the auxetic behavior of a crumpled metamaterial manufactured with aluminum foil, so far such experimental evidence has not been reported by other authors. Thus, 2D digital image correlation (DIC) was used to make full field measurements of longitudinal and transverse strains on the surface of crumpled aluminum specimens under axial compression; it was no possible to use conventional methods on the measurements of strains, such as strain gages or extensometers, due to the irregular topography on the surface of the specimens. The obtained results exhibit the auxetic nature of the aluminum crumpled material, with Poisson´s ratio values that vary from ?=-0.17 to ?=-0.02 for relative densities in the range ?_r=0.07 to ?_r=0.21. Keywords: Auxetic, Digital Image Correlation, Crumpled material, Poisson, Computer vision system.

Efficacy of a computer vision-based system for exercise management in patients with knee osteoarthritis: a study protocol for a randomised controlled pilot trial

IntroductionThis study aims to evaluate the efficacy of a computer vision system in guiding exercise management for patients with knee osteoarthritis (OA) by comparing functional improvement between a tele-rehabilitation versus...

Seam estimation based on dense matching for parallax-tolerant image stitching

Image stitching with large parallax poses a significant challenge in the field of computer vision. Existing seam-based approaches attempt to address parallax artifacts by stitching images along seams. However, issues such as object mismatches, disappearances, and duplications still arise occasionally, primarily due to inaccurate alignment of dense pixels or inappropriate seam estimation methods. In this paper, we propose a robust seam-based parallax-tolerant image stitching method that leverages dense flow estimation from state-of-the-art approaches. Firstly, we develop a seam estimation method that does not require pre-estimation of image warping model. Instead, it directly estimates the seam by measuring the local smoothness of the optical flow field and incorporating a penalty term for duplications. Subsequently, we design an iterative algorithm that utilizes the location of estimated seam to solve a spatial smooth warping model and eliminate outlier corresponding pairs. By employing this approach, we effectively address the intertwined challenges of estimating the warping model and seam. Experiment on real-world images shows that our proposed method achieves superior local alignment accuracy near the stitching seam and outperforms other state-of-the-art techniques on visual stitching result.

GradToken: Decoupling tokens with class-aware gradient for visual explanation of Transformer network

Transformer networks have been widely used in the fields of computer vision, natural language processing, graph-structured data analysis, etc. Subsequently, explanations of Transformer play a key role in helping humans understand and analyze its decision-making and working mechanism, thereby improving the trustworthiness in its real-world applications. However, it is difficult to apply the existing explanation methods for convolutional neural networks to Transformer networks, due to the significant differences between their structures. How to design a specific and effective explanation method for Transformer poses a challenge in the explanation area. To address this challenge, we first analyze the semantic coupling problem of attention weight matrices in Transformer, which puts obstacles in providing distinctive explanations for different categories of targets. Then, we propose a gradient-decoupling-based token relevance method (i.e., GradToken) for the visual explanation of Transformer’s predictions. GradToken exploits the class-aware gradient to decouple the tangled semantics in the class token to the semantics corresponding to each category. GradToken further leverages the relations between the class token and spatial tokens to generate relevance maps. As a result, the visual explanation results generated by GradToken can effectively focus on the regions of selected targets. Extensive quantitative and qualitative experiments are conducted to verify the validity and reliability of the proposed method.

Does concomitant diazepam and ethanol use modulate age-related cognitive decline in mice?

BackgroundConcomitant use of alcohol and benzodiazepines are described among elderly, raising concerns about their combined impact on memory. We aimed to evaluate the long-term impact of chronic diazepam use associated with ethanol intoxication on memory in aging mice. MethodsTwelve-month-old male C57BL6 mice were assigned into 4 groups: ethanol (OH), diazepam (DIA), diazepam + ethanol (DOH) and control (CTL). For 16 weeks, ethanol was available ad libitum and diazepam was mixed with food. Behavioral testing, performed during and after treatment cessation included working memory and visual recognition memory assessment. The second session was implemented with spatial reference learning and memory assessment in the Barnes maze test. In vivo magnetic resonance spectroscopy (MRS) acquisitions were performed to quantify hippocampal metabolites during and after cessation treatment. ResultsDuring treatment, visual recognition memory was significantly different between groups with the DIA group exhibiting the worst performance. MRS acquisition highlighted higher glutamate and choline levels in OH and DOH groups in comparison to CTL and DIA groups. After treatment wash-out, there was no difference between in the different memories evaluated. Only the learning phase of the spatial reference memory test differed significantly with worst performance in OH groups. Three months after treatment cessation, there was no remanent effect of diazepam + ethanol on hippocampal metabolites changes. ConclusionsWe did not evidence additive effect of ethanol and diazepam on memory and hippocampal metabolite levels. The disturbances observed during treatment were no remanent, highlighting the benefits of discontinuing these substances.

Multimodal Strain Sensing System for Shape Recognition of Tensegrity Structures by Combining Traditional Regression and Deep Learning Approaches

Multimodal Strain Sensing System for Shape Recognition of Tensegrity Structures by Combining Traditional Regression and Deep Learning Approaches

User emotion recognition and indoor space interaction design: a CNN model optimized by multimodal weighted networks

In interior interaction design, achieving intelligent user-interior interaction is contingent upon understanding the user’s emotional responses. Precise identification of the user’s visual emotions holds paramount importance. Current visual emotion recognition methods rely solely on singular features, predominantly facial expressions, resulting in inadequate coverage of visual characteristics and low recognition rates. This study introduces a deep learning-based multimodal weighting network model to address this challenge. The model initiates with a convolutional attention module, employing a self-attention mechanism within a convolutional neural network (CNN). As a result, the multimodal weighting network model is integrated to optimize weights during training. Finally, a weight network classifier is derived from these optimized weights to facilitate visual emotion recognition. Experimental outcomes reveal a 77.057% correctness rate and a 74.75% accuracy rate in visual emotion recognition. Comparative analysis against existing models demonstrates the superiority of the multimodal weight network model, showcasing its potential to enhance human-centric and intelligent indoor interaction design.

Modeling the Effect of Greenways’ Multilevel Visual Characteristics on Thermal Perception in Summer Based on Bayesian Network and Computer Vision

The aim of this study is to reveal the effects of multilevel visual characteristics of greenways on thermal perception in hot and humid regions during summer and to explore the potential of visual design to enhance psychological thermal comfort. Data on light (L), color (C), plant richness (PR), space openness (SO), scenic view (SV), thermal sensation (TS), and thermal preference (TP) were collected through questionnaires (n = 546). Computer vision technology was applied to measure the green view index (GVI), sky view index (SVI), paving index (PI), spatial enclosure (SE), and water index (WI). Using the hill climbing algorithm in R to construct a Bayesian network, model validation results indicated prediction accuracies of 0.799 for TS and 0.838 for TP. The results showed that: (1) SE, WI, and SV significantly positively influence TS, while L significantly negatively influences TS (R2 = 0.6805, p-value < 0.05); (2) WI, TS, and SV significantly positively influence TP (R2 = 0.759, p-value < 0.05).

Exploring the Implementation Methods of AI-Driven Robots in Media Interaction Experience Design

The purpose of this study is to deeply explore the innovative implementation methods of artificial intelligence (AI)-driven robots in media interactive experience design. Firstly, this paper expounds how AI-driven robot technology can provide users with a more natural and intelligent interactive experience through natural language processing (NLP), speech recognition, computer vision and other technologies, thus enhancing users' sense of participation and immersion. In order to realize the efficient application of AI-driven robot in media interaction, this study proposes a series of implementation methods, including improving the accuracy and personalization of robot interaction by using NLP and deep learning algorithm; Identify user expressions and gestures through computer vision technology to enhance the naturalness of interaction; Multi-modal interactive technology is used to integrate multi-sensory channel information to provide comprehensive interactive experience; Use deep learning technology to train the robot's ability to perceive and understand users' needs, and introduce emotion analysis technology to realize more humanized service; By constructing user portraits, we design personalized interaction schemes that adapt to different scenarios and users. In the experimental part, three different scenes, library, shopping mall and museum, are selected to verify the effectiveness of the design scheme. The results show that the AI-driven robot significantly improves the user's satisfaction and emotional experience under the personalized interaction scheme.

ACMamba: A State Space Model-Based Approach for Multi-Weather Degraded Image Restoration

In computer vision, eliminating the effects of adverse weather conditions such as rain, snow, and fog on images is a key research challenge. Existing studies primarily focus on image restoration for single weather types, while methods addressing image restoration under multiple combined weather conditions remain relatively scarce. Furthermore, current mainstream restoration networks, mostly based on Transformer and CNN architectures, struggle to achieve an effective balance between global receptive field and computational efficiency, limiting their performance in practical applications. This study proposes ACMamba, an end-to-end lightweight network based on selective state space models, aimed at achieving image restoration under multiple weather conditions using a unified set of parameters. Specifically, we design a novel Visual State Space Module (VSSM) and a Spatially Aware Feed-Forward Network (SAFN), which organically combine the local feature extraction capabilities of convolutions with the long-range dependency modeling capabilities of selective state space models (SSMs). This combination significantly improves computational efficiency while maintaining a global receptive field, enabling effective application of the Mamba architecture to multi-weather image restoration tasks. Comprehensive experiments demonstrate that our proposed approach significantly outperforms existing methods for both specific and multi-weather tasks across multiple benchmark datasets, showcasing its efficient long-range modeling potential in multi-weather image restoration tasks.

Autonomous Drone Navigation Using Computer Vision

Abstract: Autonomous drone navigation with computer vision is an innovative technology that allows drones to move through intricate surroundings without needing human control. This system uses live visual information to identify objects, steer clear of obstacles, and determine routes, improving operational safety and precision. The project's main objective is to create a visionbased navigation system that combines object detection and obstacle avoidance algorithms through deep learning methods like YOLO (You Only Look Once), alongside real-time sensor fusion. The drone uses computer vision algorithms to process aerial images and automatically changes its flight path to prevent crashes. A personalized dataset of aerial images is generated and utilized for improving object detection skills in various environments. In order to guarantee practicality in real- world situations, the system undergoes validation through simulations and field tests on different terrains, focusing on its resilience in changing environments. Improvements in navigation accuracy and obstacle detection are accomplished by implementing adaptive pathplanning and integrating multiple sensors, guaranteeing the drone's efficient operation in real-life situations. The goal of this method is to enhance the drone's ability to make decisions, minimize human mistakes, and expand its range of potential uses in areas like surveillance, agriculture, and disaster response

Waste Segregator: Waste Management Prototype

Smart Waste Dustbins utilize IoT, AI, and computer vision for efficient waste management. They feature multiple compartments, sensors, and automatic lid mechanisms for real-time waste sorting. Computer vision and AI algorithms accurately identify and segregate waste. Smart Waste Dustbins monitor fill levels and offer a user-friendly interface for public engagement. Sensors include ultrasonic, moisture, and metal sensors for proximity, moisture analysis, and metal item segregation. Smart Waste Dustbins revolutionize waste management, enhancing efficiency in segregation, recycling, and collection processes, leading to reduced environmental impact and improved resource utilization for sustainable communities.

Advances in Markerless Motion Capture Systems: A Review of OpenCap and Its Applications

Motion capture technology has long been a cornerstone of biomechanical research and analysis, traditionally relying on marker-based systems. However, these systems have inherent limitations, including high costs, time-consuming setup, and constraints on natural movement. Recent advancements in computer vision and machine learning have paved the way for markerless motion capture systems, such as OpenCap, which offer a more accessible and natural approach to biomechanical analysis. This review focuses on OpenCap, exploring its accuracy, challenges, and future potential, while comparing it with traditional motion capture technologies and discussing its applications in sports, clinical rehabilitation, and everyday use. Areas in need of refinement, such as improving pose estimation algorithms and addressing inter-trial variability, are identified as key future research directions. Other markerless systems are also compared in terms of advantages, limitations, and application. The findings suggest that while OpenCap and similar systems hold significant promise, further research and refinement are necessary to fully realize their potential and integrate them seamlessly into various biomechanical applications.

Random-Coupled Neural Network

Improving the efficiency of current neural networks and modeling them on biological neural systems have become prominent research directions in recent years. The pulse-coupled neural network (PCNN) is widely used to mimic the computational characteristics of the human brain in computer vision and neural network fields. However, PCNN faces limitations such as limited neural connections, high computational costs, and a lack of stochastic properties. This study proposes a random-coupled neural network (RCNN) to address these limitations. RCNN employs a stochastic inactivation process, selectively inactivating neural connections using a random inactivation weight matrix. This method reduces the computational burden and allows for extensive neural connections. RCNN encodes constant stimuli as periodic spike trains and periodic stimuli as chaotic spike trains, reflecting the information encoding characteristics of biological neural systems. Our experiments applied RCNN to image segmentation and fusion tasks, demonstrating its robustness, efficiency, and high noise resistance. Results indicate that RCNN surpasses traditional methods in performance across these applications.

Artificial Intelligence Scene Creation with Virtual Technology for the DigitalFilm

Artificial Intelligence-based scene creation leverages advanced algorithms to generate realistic and immersive environments for various applications, including gaming, virtual reality, and simulations. By utilizing techniques such as procedural generation, deep learning, and computer vision, these systems can automatically create complex landscapes, detailed textures, and dynamic elements that respond to user interactions. This not only enhances the user experience but also significantly reduces the time and effort required for manual scene design, enabling developers to focus on creativity and innovation. In the realm of digital media production, the quest for lifelike scenes and efficient evaluation methodologies has led to the exploration of novel techniques. This paper investigates the synergy between point estimation and artificial intelligence (AI) in advancing digital scene creation and evaluation. Through comprehensive simulations and analyses, we assess the effectiveness of point estimation in quantifying scene attributes such as visual realism, dynamic interactivity, physical accuracy, and artistic expression. Additionally, we delve into the utilization of AI algorithms for automating scene classification, thereby streamlining decision-making processes and optimizing resource allocation in production workflows. Through comprehensive simulations and analyses, we assess the effectiveness of point estimation in quantifying scene attributes such as visual realism (mean score: 0.85), dynamic interactivity (mean score: 0.72), physical accuracy (mean score: 0.93), and artistic expression (mean score: 0.65).

Parallel and High Throughput Reaction Monitoring with Computer Vision

AbstractWe report the development and applications of a computer vision based reaction monitoring method for parallel and high throughput experimentation (HTE). Whereas previous efforts reported methods to extract bulk kinetics of one reaction from one video, this new approach enables one video to capture bulk kinetics of multiple reactions running in parallel. Case studies, in and beyond well‐plate high throughput settings, are described. Analysis of parallel dye‐quenching hydroxylations, DMAP‐catalysed esterification, solid‐liquid sedimentation dynamics, metal catalyst degradation, and biologically‐relevant sugar‐mediated nitro reduction reactions have each provided insight into the scope and limitations of camera‐enabled high throughput kinetics as a means of widening known analytical bottlenecks in HTE for reaction discovery, mechanistic understanding, and optimisation. It is envisaged that the nature of the multi‐reaction time‐resolved datasets made available by this analytical approach will later serve a broad range of downstream efforts in machine learning approaches to exploring chemical space.

PicPatrolling: Smart Surveillance with ESP32CAM

Robotics is an exciting and developing branch of science. Robots are now common in society because of their high efficiency and durability, and they have many users. Security and surveillance robots, especially those with wheeled movements and agility, are important for multiplexes in security and surveillance domains, including public security, military, facilities, and home security. These robots carry built-in high-end sensors, cameras, and communication equipment that simplify real-time data collection in indoor and outdoor environments. "PicPatrolling: Smart Surveillance with ESP32CAM” is a technique that attempts to improve the current concept of security and surveillance. The proposed system incorporates ESP32CAM for image capture and transmission without delay and GPS for location tracking. The system includes sending photo e-mail alerts, making immediate reports, and guaranteeing adequate area coverage. The robot can move during operation in a complex environment using computer vision, machine learning, and sensor fusion to recognize objects, faces, and changes and monitor important regions.. Application areas for surveillance robots include commercial and populated regions, transport points and architectural structures, other soft regions that people cannot access, and other dangerous areas that humans cannot monitor for various reasons.

A Picture May Be Worth a Hundred Words for Visual Question Answering

How far can textual representations go in understanding images? In image understanding, effective representations are essential. Deep visual features from object recognition models currently dominate various tasks, especially Visual Question Answering (VQA). However, these conventional features often struggle to capture image details in ways that match human understanding, and their decision processes lack interpretability. Meanwhile, the recent progress in language models suggests that descriptive text could offer a viable alternative. This paper investigated the use of descriptive text as an alternative to deep visual features in VQA. We propose to process description–question pairs rather than visual features, utilizing a language-only Transformer model. We also explored data augmentation strategies to enhance training set diversity and mitigate statistical bias. Extensive evaluation shows that textual representations using approximately a hundred words can effectively compete with deep visual features on both the VQA 2.0 and VQA-CP v2 datasets. Our qualitative experiments further reveal that these textual representations enable clearer investigation of VQA model decision processes, thereby improving interpretability.