Problem In Computer Vision Research Articles

Data augmentation based face anti-spoofing (FAS) scheme using deep learning techniques

Recent years have seen a rise in interest in face anti-spoofing (FAS) owing to the critical function it plays in protecting face recognition systems against presentation assaults (PAs). Early-stage FAS approaches relying on handmade characteristics become inaccurate when steadily realistic PAs of unique sorts emerge. Thus, face anti-spoofing algorithms are gaining increasing relevance in such setups. A very innovative method called deep learning has shown remarkable success in difficult computer vision problems. The proposed method uses deep acquisition and transfer of learning to extract characteristics from people’s faces. This is why the authors of this study recommend using the Faster RCNN classifier with a face-liveness detection approach. Two distinct components— the data augmentation module for assessing sparse information as well as the faster RCNN classifier module— make up the anti-spoofing approach. We may use any publicly accessible dataset to train our quicker RCNN classifier. We successively fused these two components and used the Android platform to create a basic face recognition app. The results of the tests demonstrate that the developed module can identify several types of face spoof assaults, such as those carried out with the use of posters, masks, or cell phones. Testing the proposed architecture both across and inside databases using three benchmarking (Idiap Replay Attack, CASIA- FASD, & 3DMAD) demonstrate its ability to deliver outcomes on par with cutting-edge techniques.

An Enhanced Deep Learning Model for Brain Tumor Prediction

Brain tumour diagnosis & prediction is an challenging issue and important area of research. perversely, convolutional neural networks can support this (CNNs). They have mastered computer vision problems as well as other issues like segmenting, detecting, and recognizing visual objects. By enhancing the brain images with help of segmentation methods that are extremely challenging related to noise and cluster size sensitivity issues, as well as automated region of Interest detection (ROI), it helps with the diagnosis of brain tumours. The reality that CNNs have achieved high level of accuracy and it does not require manual extraction of features. Finding a brain tumour and correctly classifying it are challenging tasks. CNN outperforms rivals due to its extensive use in image recognition. Brain tumour segmentation is the most significant and challenging problems in the field of medical image processing research because human assisted manual categorization may lead to inaccurate prediction and diagnosis. In addition, when there is a huge amount of data existing to support in the process, it is challenging. Extraction of tumour areas from images becomes challenging due to the wide variety of appearances of brain tumours and the similarity of tumour and normal tissues.

Rancang Bangun Kunci Pintu Perpustakaan Otomatis Menggunakan Algoritma Histogram of oriented gradients Berbasis Mini-PC

Facial recognition has been a long-standing problem in computer vision. Recently, Histograms of Oriented Gradients (HOGs) have been shown to be effective descriptors for object recognition in general and face recognition in particular. In this study, HOGs are used as an algorithm to achieve facial recognition, which aims to recognize the faces of registered students when visiting the library without the need to fill in the attendance list in the visitor book and open the door and lock the door automatically.

Local-to-Global Panorama Inpainting for Locale-Aware Indoor Lighting Prediction.

Predicting panoramic indoor lighting from a single perspective image is a fundamental but highly ill-posed problem in computer vision and graphics. To achieve locale-aware and robust prediction, this problem can be decomposed into three sub-tasks: depth-based image warping, panorama inpainting and high-dynamic-range (HDR) reconstruction, among which the success of panorama inpainting plays a key role. Recent methods mostly rely on convolutional neural networks (CNNs) to fill the missing contents in the warped panorama. However, they usually achieve suboptimal performance since the missing contents occupy a very large portion in the panoramic space while CNNs are plagued by limited receptive fields. The spatially-varying distortion in the spherical signals further increases the difficulty for conventional CNNs. To address these issues, we propose a local-to-global strategy for large-scale panorama inpainting. In our method, a depth-guided local inpainting is first applied on the warped panorama to fill small but dense holes. Then, a transformer-based network, dubbed PanoTransformer, is designed to hallucinate reasonable global structures in the large holes. To avoid distortion, we further employ cubemap projection in our design of PanoTransformer. The high-quality panorama recovered at any locale helps us to capture spatially-varying indoor illumination with physically-plausible global structures and fine details.

Investigating transformer‐based models for spatial downscaling and correcting biases of near‐surface temperature and wind‐speed forecasts

AbstractHigh‐resolution and accurate prediction of near‐surface weather parameters based on numerical weather prediction (NWP) models is essential for many downstream and real‐world applications. Traditional dynamical or statistical downscaling methods are insufficient to derive high‐resolution data from operational NWP forecasts, making it essential to devise new approaches. In recent years, an increasing number of researchers have explored the implementations of deep learning (DL) based models for spatial downscaling, motivated by the similarity between the super‐resolution (SR) problem in computer vision (CV) and downscaling. Furthermore, while transformer‐based models have become state‐of‐the‐art models for many SR tasks, they are rarely applied for downscaling of weather forecasts or climate projections. This study adapted transformer‐based models such as SwinIR and Uformer to downscale the temperature at 2 m () and wind speed at 10 m () over Eastern Inner Mongolia, encompassing the area from 39.6–46°N latitude and 111.6–118°E longitude. We used high‐resolution forecast (HRES) data from the European Centre for Medium‐range Weather Forecast (ECMWF) with a spatial resolution of 0.1° as the input and gridded observation data from the China Meteorological Administration (CMA) Land Data Assimilation System (CLDAS) at a spatial resolution of 0.01° as the target. Given that the models use observation data rather than a coarse‐grained version of forecast data as the target, they accomplish both bias correction and spatial downscaling. The results demonstrate that the performance of SwinIR and Uformer is superior to that of two convolutional neural network (CNN) based models (UNet and RCAN). Additionally, we introduced a novel module to extract features of varying resolution from the high‐resolution topography data and applied a multiscale feature fusion module to merge features of different scales, contributing to further enhancement of Uformer's performance.

Analysis of transformer-based models for time series data, natural language processing, and computer vision

The birth of the Transformer revolutionarily signalled the start of a new epic chapter in the deep learning era. Through an encoder-decoder architecture, including residual connection, multi-head self-attention, etc., it completely reformed the deep models and unified the models used in traditional computer vision (CV) and natural language processing (NLP) problems. In recent years, many papers published have adapted the original Transformer model to better complete tasks in time series analysis, CV, and NLP. In the area of natural language processing, Bidirectional Encoder Representations from Transformers (BERT) employs a two-way transformer structure to learn context-based language representation, whereas Generative Pre-trained Transformer (GPT) employs a one-way transformer but enhances corpus training to enhance the model effect. The Vision Transformer model is the cornerstone of computer vision. It separates the input image into various patches, projects each patch into vectorized features, and then passes the them to Transformer. Based on the idea of the Vision Transformer, Swin Transformer and Biformer further optimized the Transformer and achieved better results. Time series combines the ideas embodied in CV and NLP, and in doing so, improves the specificity and various difficulties of time series problems to lower algorithm complexity and increase prediction accuracy. This article summarizes the uses and improvements of the Transformer in NLP, CV and time series, explores the development history and ideas on algorithm optimization, and predicts the potential developments of Transformer in these three fields.

TIC: text-guided image colorization using conditional generative model

Image colorization is a well-known problem in computer vision. However, due to the ill-posed nature of the task, image colorization is inherently challenging. Though several attempts have been made by researchers to make the colorization pipeline automatic, these processes often produce unrealistic results due to a lack of conditioning. In this work, we attempt to integrate textual descriptions as an auxiliary condition, along with the grayscale image that is to be colorized, to improve the fidelity of the colorization process. To the best of our knowledge, this is one of the first attempts to incorporate textual conditioning in the colorization pipeline. To do so, a novel deep network has been proposed that takes two inputs (the grayscale image and the respective encoded text description) and tries to predict the relevant color gamut. As the respective textual descriptions contain color information of the objects present in the scene, the text encoding helps to improve the overall quality of the predicted colors. The proposed model has been evaluated using different metrics like SSIM, PSNR, LPISPS and achieved scores of 0.917, 23.27,0.223, respectively. These quantitative metrics have shown that the proposed method outperforms the SOTA techniques in most of the cases.

Semantic visual simultaneous localization and mapping (SLAM) using deep learning for dynamic scenes.

Simultaneous localization and mapping (SLAM) is a fundamental problem in robotics and computer vision. It involves the task of a robot or an autonomous system navigating an unknown environment, simultaneously creating a map of the surroundings, and accurately estimating its position within that map. While significant progress has been made in SLAM over the years, challenges still need to be addressed. One prominent issue is robustness and accuracy in dynamic environments, which can cause uncertainties and errors in the estimation process. Traditional methods using temporal information to differentiate static and dynamic objects have limitations in accuracy and applicability. Nowadays, many research trends have leaned towards utilizing deep learning-based methods which leverage the capabilities to handle dynamic objects, semantic segmentation, and motion estimation, aiming to improve accuracy and adaptability in complex scenes. This article proposed an approach to enhance monocular visual odometry's robustness and precision in dynamic environments. An enhanced algorithm using the semantic segmentation algorithm DeeplabV3+ is used to identify dynamic objects in the image and then apply the motion consistency check to remove feature points belonging to dynamic objects. The remaining static feature points are then used for feature matching and pose estimation based on ORB-SLAM2 using the Technical University of Munich (TUM) dataset. Experimental results show that our method outperforms traditional visual odometry methods in accuracy and robustness, especially in dynamic environments. By eliminating the influence of moving objects, our method improves the accuracy and robustness of visual odometry in dynamic environments. Compared to the traditional ORB-SLAM2, the results show that the system significantly reduces the absolute trajectory error and the relative pose error in dynamic scenes. Our approach has significantly improved the accuracy and robustness of the SLAM system's pose estimation.

Evaluation of Transfer Learning Techniques for Fault Classification in Radial Distribution Systems: A Comparative Study

Transfer learning has recently had a detectable impact on the state of the art in a wide variety of applications, and this trend is expected to continue in the near future. Both transfer learning and deep learning algorithms make use of a number of network layers, each of which may be intellectually learned and typically represents the data in a hierarchical fashion with increasing levels of abstraction. Convolutional neural networks have been proven to be exceptionally successful machine learning and deep learning techniques for a number of computer vision problems. These networks were developed by companies such as Alexa, Google, and Squeeze. Fault diagnostic strategies that are based on deep learning techniques are currently a topic of intense investigation due to their higher performance. Using transfer learning technology to carry out fault categorization in a power distribution system in a manner that is both accurate and efficient The work at hand employs a fault classification model for a radial power distribution system that is based on transfer learning and deep learning. Images of time series of three-phase fault currents are acquired via simulation with the assistance of PSCAD software as part of the proposed approach for doing so. In the next step, CNN models that are based on Alex Net, Google Net, and Squeeze Net are utilized to extract fault features from defective photos in order to categorize eleven distinct defects (using the MATLAB platform). For the categorization of defects in a radial distribution system, Alex Net, Google Net, and SqueezNet each offer accuracy of approximately 98.92%, 97.48%, and 99.82%, respectively. In this study, the classification of faults in a distribution system is accomplished with the help of AlexNet, GoogleNet, and SqueezNet. According to the findings of the simulations, the test accuracy for SqueezeNet is the highest it can be, coming in at 99.82%. Because of this, selecting it as the solution to the issue of fault classification in the test distribution system is your best option.

4TH INDUSTRIAL REVOLUTION AND ARTIFICIAL INTELLIGENCE

An overview of industrial revolutions is made. Then the analysis of the fourth industrial revolution. The main major components of Industry 4.0 are identified, and their essence is revealed. The tasks of artificial intelligence within Industry 4.0 are also considered. Developed software in Python language about data and machine learning (ML) to visualize objects and table data. In order to use the capabilities of AI to the maximum benefit for business, it is necessary to hire data scientists. Data science combines statistics, computer science and business knowledge to extract value from various data sources. Developers use artificial intelligence to more effectively perform tasks that would otherwise have to be done manually, interact with customers, identify patterns and solve problems. To start working with AI, developers will need mathematical knowledge and the ability to use algorithms. There are several stages in the development and deployment of machine learning models, including training and inference. AI learning and inference refers to the process of experimenting with machine learning models to solve a problem. For example, a machine learning engineer can experiment with various candidate models to solve a computer vision problem, such as detecting bone fractures in X-ray images. Keywords: fourth industrial revolution, artificial intelligence.

Face deblurring based on regularized structure and enhanced texture information

Image deblurring is an essential problem in computer vision. Due to highly structured and special facial components (e.g. eyes), most general image deblurring methods and face deblurring methods failed to yield more explicit structure and facial details, resulting in too smooth, uncoordinated and distorted face structure. Considering the unique face texture and sufficient facial details, we present an effective face deblurring network by exploiting more regularized structure and enhanced texture information (RSETNet). We first incorporate the face parsing network with fine-tuning to obtain more accurate face structure, and we present the feature adaptive denormalization (FAD) to regularize the facial structure as a condition of auxiliary to generate more harmonious and undistorted face structure. Meanwhile, to improve the generated facial texture information, we propose a new Laplace depth-wise separable convolution (LDConv) and multi-patch discriminator. Compared with existing methods, our face deblurring method could restore face structure more accurately and with more facial details. Experiments on two public face datasets have demonstrated the effectiveness of our proposed methods in terms of qualitative and quantitative indicators.

Monte Carlo DropBlock for modeling uncertainty in object detection

With the advancements made in deep learning, computer vision problems have seen a great improvement in performance. However, in many real-world applications such as autonomous driving vehicles, the risk associated with incorrect predictions of objects or segmentation of images is very high. Standard deep learning models for object detection and segmentation such as YOLO models are often overconfident in their predictions and do not take into account the uncertainty in predictions on out-of-distribution data. In this work, we propose an efficient and effective approach, Monte-Carlo DropBlock (MC-DropBlock), to model uncertainty in YOLO and convolutional vision Transformers for object detection. The proposed approach applies drop-block during training time and testing time on the convolutional layer of the deep learning models such as YOLO and convolutional transformer. We theoretically show that this leads to a Bayesian convolutional neural network capable of capturing the epistemic uncertainty in the model. Additionally, we capture the aleatoric uncertainty in the data using a Gaussian likelihood. We demonstrate the effectiveness of the proposed approach on modeling uncertainty in object detection and segmentation tasks using out-of-distribution experiments. Experimental results show that MC-DropBlock improves the generalization, calibration, and uncertainty modeling capabilities of YOLO models and convolutional Transformer models for object detection and segmentation.

Generalization of Neural Networks on Second-Order Hypercomplex Numbers

The vast majority of existing neural networks operate by rules set within the algebra of real numbers. However, as theoretical understanding of the fundamentals of neural networks and their practical applications grow stronger, new problems arise, which require going beyond such algebra. Various tasks come to light when the original data naturally have complex-valued formats. This situation is encouraging researchers to explore whether neural networks based on complex numbers can provide benefits over the ones limited to real numbers. Multiple recent works have been dedicated to developing the architecture and building blocks of complex-valued neural networks. In this paper, we generalize models by considering other types of hypercomplex numbers of the second order: dual and double numbers. We developed basic operators for these algebras, such as convolution, activation functions, and batch normalization, and rebuilt several real-valued networks to use them with these new algebras. We developed a general methodology for dual and double-valued gradient calculations based on Wirtinger derivatives for complex-valued functions. For classical computer vision (CIFAR-10, CIFAR-100, SVHN) and signal processing (G2Net, MusicNet) classification problems, our benchmarks show that the transition to the hypercomplex domain can be helpful in reaching higher values of metrics, compared to the original real-valued models.

Scaling-Invariant Max-Filtering Enhancement Transformers for Efficient Visual Tracking

Scaling-Invariant Max-Filtering Enhancement Transformers for Efficient Visual Tracking

Efficient Dehazing with Recursive Gated Convolution in U-Net: A Novel Approach for Image Dehazing.

Image dehazing, a fundamental problem in computer vision, involves the recovery of clear visual cues from images marred by haze. Over recent years, deploying deep learning paradigms has spurred significant strides in image dehazing tasks. However, many dehazing networks aim to enhance performance by adopting intricate network architectures, complicating training, inference, and deployment procedures. This study proposes an end-to-end U-Net dehazing network model with recursive gated convolution and attention mechanisms to improve performance while maintaining a lean network structure. In our approach, we leverage an improved recursive gated convolution mechanism to substitute the original U-Net's convolution blocks with residual blocks and apply the SK fusion module to revamp the skip connection method. We designate this novel U-Net variant as the Dehaze Recursive Gated U-Net (DRGNet). Comprehensive testing across public datasets demonstrates the DRGNet's superior performance in dehazing quality, detail retrieval, and objective evaluation metrics. Ablation studies further confirm the effectiveness of the key design elements.

Causal multi-label learning for image classification

In this paper, we investigate the problem of causal image classification with multi-label learning. As multi-label learning involves a diversity of supervision signals, it is considered a challenging issue to solve. Previous approaches have attempted to improve performance by identifying label-related image areas or exploiting the co-occurrence of labels. However, these methods are often characterized by complicated procedures, tedious computations, and a lack of intuitive interpretations. To overcome these limitations, we propose a novel approach that incorporates the concept of causal inference, which has been shown to be beneficial in other computer vision problems. Our method, called causal multi-label learning (CMLL), enables the selection of multiple objects from the original image through a multi-class attention module. These objects are then subjected to causal intervention to learn the causal relationships between different labels. Our proposed approach is both elegant and effective, with low computational cost and few parameters required for the multi-class causal intervention approach. Extensive tests and ablation studies demonstrate that the proposed method significantly improves prediction performance without a significant increase in training and inference times.

Stressed Coral Reef Identification Using Deep Learning CNN Techniques

Deep learning is a machine learning technique that allows the computer to process things that occur naturally to humans. Today, deep learning techniques are commonly used in computer vision to classify images and videos. As a result, for challenging computer vision problems, deep learning provides state of the art solutions to it. Coral reefs are an essential resource of the earth. A new study finds the planet has lost half of its coral reefs since 1950. It is necessary to restore and prevent damage to coral reefs as they play an important role in maintaining a balance in the marine ecosystem. This proposed work helps to prevent the corals from bleaching and restore them to a healthy condition by identifying the root cause of the threats. In the proposed work, using deep learning CNN techniques, the images are classified into Healthy and Stressed coral reefs. Stressed coral reefs are an intermediate state of coral reef between healthy and bleached coral reefs. The pre-trained models Resnet50 and Inception V3 are used in this study to classify the images. Also, a proposed CNN model is built and tested for the same. The results of Inception V3 and Resnet50 are improved to 70% and 55% by tuning the hypermeters such as dropouts and batch normalisation. Similarly, the proposed model is tuned as required and obtains a maximum of up to 90% accuracy. With large datasets, the optimum amount of neural networks and tuning it as required brings higher accuracy than other methods.

Computer Vision Based Planogram Compliance Evaluation

Arranging products in stores according to planograms, optimized product arrangement maps, is an important sales enabler and necessary for keeping up with the highly competitive modern retail market. Key benefits of planograms include increased efficiency, maximized retail store space, increased customer satisfaction, visual appeal, and increased revenue. The planograms are realized into product arrangements by humans, a process that is prone to mistakes. Therefore, for optimal merchandising performance, the planogram compliance of the arrangements needs to be evaluated from time to time. We investigate utilizing a computer vision problem setting—retail product detection—to automate planogram compliance evaluation. Retail product detection comprises product detection and classification. The detected and classified products can be compared to the planogram in order to evaluate compliance. In this paper, we propose a novel retail product detection pipeline combining a Gaussian layer network product proposal generator and domain invariant hierarchical embedding (DIHE) classifier. We utilize the detection pipeline with RANSAC pose estimation for planogram compliance evaluation. As the existing metrics for evaluating the planogram compliance evaluation performance assume unrealistically that the test image matches the planogram, we propose a novel metric, called normalized planogram compliance error (EPC), for benchmarking real-world setups. We evaluate the performance of our method with two datasets: the only open-source dataset with planogram evaluation data, GP-180, and our own dataset collected from a large Nordic retailer. Based on the evaluation, our method provides an improved planogram compliance evaluation pipeline, with accurate product location estimation when using real-life images that include entire shelves, unlike previous research that has only used images with few products. Our analysis also demonstrates that our method requires less processing time than the state-of-the-art compliance evaluation methods.

Fast and robust for texture-less feature registration via adaptive heterogeneous kernels

Feature registration is a core problem in computer vision and machine learning techniques. In recent years, significant progress has been made in learning-based feature registration algorithms. However, such methods usually capture image-level features and strive to capture contour information, inadvertently neglecting the exploration of low-texture regions, while the complex model structure brings more parameters. Therefore, it remains a challenging task to effectively address low-texture feature registration under limited sample conditions. In this paper, we propose a fast and robust feature registration algorithm to achieve fast feature registration in the low-texture case. Specifically, we first construct a two-step learning framework and use a faster adaptive heterogeneous filter (AdaH-Filter) to guide coarse-grained and fine-grained feature extraction. Then, we propose registration feature refiners (ReFR) to achieve the fusion of two-stage features in a low-texture environment while guaranteeing the speed of the detector. Finally, to further improve the network performance, a Warp Extractor is proposed to improve the speed and portability of model inference. This strategy allows the network to perform different levels of fine extraction adaptively according to the registration requirements. Moreover, the proposed method is trained in an end-to-end manner, guided by cosine loss, without any additional supervision. Extensive experimental results on several challenging and representative feature registration datasets in the industry show that the proposed method outperforms the existing state-of-the-art methods. More importantly, it has a faster inference speed under different sensor modalities.

Temporal Pixel-Level Semantic Understanding Through the VSPW Dataset.

Scene understanding through pixel-level semantic parsing is one of the main problems in computer vision. Till now, image-based methods and datasets for scene parsing have been well explored. However, the real world is naturally dynamic instead of a static state. Thus, learning to perform video scene parsing is more practical for real-world applications. Considering that few datasets cover an extensive range of scenes and object categories with temporal pixel-level annotations, in this work, we present a large-scale video scene parsing dataset, namely VSPW (Video Scene Parsing in the Wild). To be specific, there are a total of 251,633 frames from 3,536 videos with densely pixel-wise annotations in VSPW, including a large variety of 231 scenes and 124 object categories. Besides, VSPW is densely annotated with a high frame rate of 15 f/s, and over 96% of videos from VSPW have high spatial resolutions from 720P to 4 K. To the best of our knowledge, VSPW is the first attempt to address the challenging video scene parsing task in the wild by considering diverse scenes. Based on our VSPW, we further propose Temporal Attention Blending (TAB) Networks to harness temporal context information for better pixel-level semantic understanding of videos. Extensive experiments on VSPW well demonstrate the superiority of the proposed TAB over other baseline approaches. We hope the new proposed dataset and the explorations in this work can help advance the challenging yet practical video scene parsing task in the future. Both the dataset and the code are available at www.vspwdataset.com.