Explore Research

Emotion recognition has become increasingly important in the field of Deep Learning (DL) and computer vision due to its broad applicability by using human-computer interaction (HCI) in areas such as psychology, healthcare, and entertainment. In this paper, we conduct a systematic review of facial and pose emotion recognition using DL and computer vision, analyzing and evaluating 77 papers from different sources under Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. Our review covers several topics, including the scope and purpose of the studies, the methods employed, and the used datasets. The scope of this work is to conduct a systematic review of facial and pose emotion recognition using DL methods and computer vision. The studies were categorized based on a proposed taxonomy that describes the type of expressions used for emotion detection, the testing environment, the currently relevant DL methods, and the datasets used. The taxonomy of methods in our review includes Convolutional Neural Network (CNN), Faster Region-based Convolutional Neural Network (R-CNN), Vision Transformer (ViT), and "Other NNs", which are the most commonly used models in the analyzed studies, indicating their trendiness in the field. Hybrid and augmented models are not explicitly categorized within this taxonomy, but they are still important to the field. This review offers an understanding of state-of-the-art computer vision algorithms and datasets for emotion recognition through facial expressions and body poses, allowing researchers to understand its fundamental components and trends.

Facial action unit (AU) detection is an important task in affective computing and has attracted extensive attention in the field of computer vision and artificial intelligence. Previous studies for AU detection usually encode complex regional feature representations with manually defined facial landmarks and learn to model the relationships among AUs via graph neural network. Albeit some progress has been achieved, it is still tedious for existing methods to capture the exclusive and concurrent relationships among different combinations of the facial AUs. To circumvent this issue, we proposed a new progressive multi-scale vision transformer (PMVT) to capture the complex relationships among different AUs for the wide range of expressions in a data-driven fashion. PMVT is based on the multi-scale self-attention mechanism that can flexibly attend to a sequence of image patches to encode the critical cues for AUs. Compared with previous AU detection methods, the benefits of PMVT are 2-fold: (i) PMVT does not rely on manually defined facial landmarks to extract the regional representations, and (ii) PMVT is capable of encoding facial regions with adaptive receptive fields, thus facilitating representation of different AU flexibly. Experimental results show that PMVT improves the AU detection accuracy on the popular BP4D and DISFA datasets. Compared with other state-of-the-art AU detection methods, PMVT obtains consistent improvements. Visualization results show PMVT automatically perceives the discriminative facial regions for robust AU detection.

Sensor-based remote healthcare monitoring is a promising approach for timely detection of adverse health events such as falls or infections in people living with dementia (PLwD) in the home, and reducing preventable hospital admissions. Current anomaly detection approaches in the home setting are hindered by challenges such as noisy, multivariate data, unreliability of event annotations, and heterogeneity across home settings. Inspired by the simplicity and recent applications of contrastive learning in the field of computer vision, we propose a lightweight, self-supervised, negative sample-free approach to detect anomalous events using home activity changes in PLwD. We use the contrastive loss between hourly-aggregated daily sensor data and a lower temporal resolution augmentation, to extract a noise-robust, discriminative representation of daily activity. The daily difference in representation forms the anomaly score, which is compared to the household-personalized threshold, and alerts raised to a clinical monitoring team. Attention weights from the Transformer encoder and Layer-wise Relevance Propagation support explainability. We evaluated the models on accuracy and generalizability, given a target alert rate. Our model outperformed state-of-the-art algorithms in detecting agitation and fall events for three distinct patient cohorts, with 86%(SD=4%) average recall and 92%(SD=4%) generalizability, at a target alert rate of 7%. Our novel application of contrastive frameworks is domain-agnostic and can extract salient patterns from time-series data in other remote monitoring environments.

This paper presents a system that utilizes vision transformers and multimodal feedback modules to facilitate navigation and collision avoidance for the visually impaired. By implementing vision transformers, the system achieves accurate object detection, enabling the real-time identification of objects in front of the user. Semantic segmentation and the algorithms developed in this work provide a means to generate a trajectory vector of all identified objects from the vision transformer and to detect objects that are likely to intersect with the user's walking path. Audio and vibrotactile feedback modules are integrated to convey collision warning through multimodal feedback. The dataset used to create the model was captured from both indoor and outdoor settings under different weather conditions at different times across multiple days, resulting in 27,867 photos consisting of 24 different classes. Classification results showed good performance (95% accuracy), supporting the efficacy and reliability of the proposed model. The design and control methods of the multimodal feedback modules for collision warning are also presented, while the experimental validation concerning their usability and efficiency stands as an upcoming endeavor. The demonstrated performance of the vision transformer and the presented algorithms in conjunction with the multimodal feedback modules show promising prospects of its feasibility and applicability for the navigation assistance of individuals with vision impairment.

Weed detection and classification using computer vision and deep learning techniques have emerged as crucial tools for precision agriculture, offering automated solutions for sustainable farming practices. This study presents a comprehensive approach to weed identification across multiple growth stages, addressing the challenges of detecting and classifying diverse weed species throughout their developmental cycles. We introduce two extensive datasets: the Alpha Weed Dataset (AWD) with 203,567 images and the Beta Weed Dataset (BWD) with 120,341 images, collectively documenting 16 prevalent weed species across 11 growth stages. The datasets were preprocessed using both traditional computer vision techniques and the advanced SAM-2 model, ensuring high-quality annotations with segmentation masks and precise bounding boxes. Our research evaluates several state-of-the-art object detection architectures, including DINO Transformer (with ResNet-101 and Swin backbones), Detection Transformer (DETR), EfficientNet B4, YOLO v8, and RetinaNet. Additionally, we propose a novel WeedSwin Transformer architecture specifically designed to address the unique challenges of weed detection, such as complex morphological variations and overlapping vegetation patterns. Through rigorous experimentation, WeedSwin demonstrated superior performance, achieving 0.993 ± 0.004 mAP and 0.985 mAR while maintaining practical processing speeds of 218.27 FPS, outperforming existing architectures across various metrics. The comprehensive evaluation across different growth stages reveals the robustness of our approach, particularly in detecting challenging "driver weeds" that significantly impact agricultural productivity. By providing accurate, automated weed identification capabilities, this research establishes a foundation for more efficient and environmentally sustainable weed management practices. The demonstrated success of the WeedSwin architecture, combined with our extensive temporal datasets, represents a significant advancement in agricultural computer vision, supporting the evolution of precision farming techniques while promoting reduced herbicide usage and improved crop management efficiency.

Pulmonary Fibrosis (PF) is an immedicable respiratory condition distinguished by permanent fibrotic alterations in the pulmonary tissue for which there is no cure. Hence, it is crucial to diagnose PF swiftly and precisely. The existing research on deep learning-based pulmonary fibrosis detection methods has limitations, including dataset sample sizes and a lack of standardization in data preprocessing and evaluation metrics. This study presents a comparative analysis of four vision transformers regarding their efficacy in accurately detecting and classifying patients with Pulmonary Fibrosis and their ability to localize abnormalities within Images obtained from Computerized Tomography (CT) scans. The dataset consisted of 13,486 samples selected out of 24647 from the Pulmonary Fibrosis dataset, which included both PF-positive CT and normal images that underwent preprocessing. The preprocessed images were divided into three sets: the training set, which accounted for 80% of the total pictures; the validation set, which comprised 10%; and the test set, which also consisted of 10%. The vision transformer models, including ViT, MobileViT2, ViTMSN, and BEiT were subjected to training and validation procedures, during which hyperparameters like the learning rate and batch size were fine-tuned. The overall performance of the optimized architectures has been assessed using various performance metrics to showcase the consistent performance of the fine-tuned model. Regarding performance, ViT has shown superior performance in validation and testing accuracy and loss minimization, specifically for CT images when trained at a single epoch with a tuned learning rate of 0.0001. The results were as follows: validation accuracy of 99.85%, testing accuracy of 100%, training loss of 0.0075, and validation loss of 0.0047. The experimental evaluation of the independently collected data gives empirical evidence that the optimized Vision Transformer (ViT) architecture exhibited superior performance compared to all other optimized architectures. It achieved a flawless score of 1.0 in various standard performance metrics, including Sensitivity, Specificity, Accuracy, F1-score, Precision, Recall, Mathew Correlation Coefficient (MCC), Precision-Recall Area under the Curve (AUC PR), Receiver Operating Characteristic and Area Under the Curve (ROC-AUC). Therefore, the optimized Vision Transformer (ViT) functions as a reliable diagnostic tool for the automated categorization of individuals with pulmonary fibrosis (PF) using chest computed tomography (CT) scans.

OBJECTIVE: To develop automated algorithms for the detection of posterior vitreous detachment (PVD) using OCT imaging.

DESIGN: Evaluation of a diagnostic test or technology.

SUBJECTS: Overall, 42 385 consecutive OCT images (865 volumetric OCT scans) obtained with Heidelberg Spectralis from 865 eyes from 464 patients at an academic retina clinic between October 2020 and December 2021 were retrospectively reviewed.

METHODS: We developed a customized computer vision algorithm based on image filtering and edge detection to detect the posterior vitreous cortex for the determination of PVD status. A second deep learning (DL) image classification model based on convolutional neural networks and ResNet-50 architecture was also trained to identify PVD status from OCT images. The training dataset consisted of 674 OCT volume scans (33 026 OCT images), while the validation testing set consisted of 73 OCT volume scans (3577 OCT images). Overall, 118 OCT volume scans (5782 OCT images) were used as a separate external testing dataset.

MAIN OUTCOME MEASURES: Accuracy, sensitivity, specificity, F1-scores, and area under the receiver operator characteristic curves (AUROCs) were measured to assess the performance of the automated algorithms.

RESULTS: Both the customized computer vision algorithm and DL model results were largely in agreement with the PVD status labeled by trained graders. The DL approach achieved an accuracy of 90.7% and an F1-score of 0.932 with a sensitivity of 100% and a specificity of 74.5% for PVD detection from an OCT volume scan. The AUROC was 89% at the image level and 96% at the volume level for the DL model. The customized computer vision algorithm attained an accuracy of 89.5% and an F1-score of 0.912 with a sensitivity of 91.9% and a specificity of 86.1% on the same task.

CONCLUSIONS: Both the computer vision algorithm and the DL model applied on OCT imaging enabled reliable detection of PVD status, demonstrating the potential for OCT-based automated PVD status classification to assist with vitreoretinal surgical planning.

FINANCIAL DISCLOSURES: Proprietary or commercial disclosure may be found after the references.

Railway track defects pose significant safety risks and can lead to accidents, economic losses, and loss of life. Traditional manual inspection methods are either time-consuming, costly, or prone to human error. This paper proposes RailTrack-DaViT, a novel vision transformer-based approach for railway track defect classification. By leveraging the Dual Attention Vision Transformer (DaViT) architecture, RailTrack-DaViT effectively captures both global and local information, enabling accurate defect detection. The model is trained and evaluated on multiple datasets including rail, fastener and fishplate, multi-faults, and ThaiRailTrack. A comprehensive analysis of the model's performance is provided including confusion matrices, training visualizations, and classification metrics. RailTrack-DaViT demonstrates superior performance compared to state-of-the-art CNN-based methods, achieving the highest accuracies: 96.9% on the rail dataset, 98.9% on the fastener and fishplate dataset, and 98.8% on the multi-faults dataset. Moreover, RailTrack-DaViT outperforms baselines on the ThaiRailTrack dataset with 99.2% accuracy, quickly adapts to unseen images, and shows better model stability during fine-tuning. This capability can significantly reduce time consumption when applying the model to novel datasets in practical applications.

This paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities and the high resolution of pixels in images compared to words in text. To address these differences, we propose a hierarchical Transformer whose representation is computed with Shifted windows. The shifted windowing scheme brings greater efficiency by limiting self-attention computation to non-overlapping local windows while also allowing for cross-window connection. This hierarchical architecture has the flexibility to model at various scales and has linear computational complexity with respect to image size. These qualities of Swin Transformer make it compatible with a broad range of vision tasks, including image classification (87.3 top-1 accuracy on ImageNet-1K) and dense prediction tasks such as object detection (58.7 box AP and 51.1 mask AP on COCO test-dev) and semantic segmentation (53.5 mIoU on ADE20K val). Its performance surpasses the previous state-of-the-art by a large margin of +2.7 box AP and +2.6 mask AP on COCO, and +3.2 mIoU on ADE20K, demonstrating the potential of Transformer-based models as vision backbones. The hierarchical design and the shifted window approach also prove beneficial for all-MLP architectures. The code and models are publicly available at https://github.com/microsoft/Swin-Transformer.

In order to develop unattended intelligent substations, numerous automatic meter reading methods have been proposed with the invention of inspection robots. However, most methods have strict restrictions on the captured image quality, so they cannot meet the application requirements of the substation. In this paper, a versatile pointer meter detection and recognition system based on computer vision is proposed for different environments. A Faster Region-based Convolutional Network (Faster R-CNN) is first used to detect the position of the target meter and the camera is adjusted according to the detection box. Then, thanks to the feature correspondence algorithm and the Perspective Transform, the problem of specular reflection and image distortion is solved to obtain high quality images. Finally, reading can be obtained after the position of pointer detected by Hough Transform. The experimental results verify the stability and accuracy of recognition system which is proved to work well under different conditions.

This paper presents a new vision Transformer, called Swin Transformer, that capably serves as a general-purpose backbone for computer vision. Challenges in adapting Transformer from language to vision arise from differences between the two domains, such as large variations in the scale of visual entities and the high resolution of pixels in images compared to words in text. To address these differences, we propose a hierarchical Transformer whose representation is computed with \textbf{S}hifted \textbf{win}dows. The shifted windowing scheme brings greater efficiency by limiting self-attention computation to non-overlapping local windows while also allowing for cross-window connection. This hierarchical architecture has the flexibility to model at various scales and has linear computational complexity with respect to image size. These qualities of Swin Transformer make it compatible with a broad range of vision tasks, including image classification (87.3 top-1 accuracy on ImageNet-1K) and dense prediction tasks such as object detection (58.7 box AP and 51.1 mask AP on COCO test-dev) and semantic segmentation (53.5 mIoU on ADE20K val). Its performance surpasses the previous state-of-the-art by a large margin of +2.7 box AP and +2.6 mask AP on COCO, and +3.2 mIoU on ADE20K, demonstrating the potential of Transformer-based models as vision backbones. The hierarchical design and the shifted window approach also prove beneficial for all-MLP architectures. The code and models are publicly available at~\url{https://github.com/microsoft/Swin-Transformer}.

Although convolutional neural networks (CNNs) have achieved great success in computer vision, this work investigates a simpler, convolution-free backbone network use-fid for many dense prediction tasks. Unlike the recently-proposed Vision Transformer (ViT) that was designed for image classification specifically, we introduce the Pyramid Vision Transformer (PVT), which overcomes the difficulties of porting Transformer to various dense prediction tasks. PVT has several merits compared to current state of the arts. (1) Different from ViT that typically yields low-resolution outputs and incurs high computational and memory costs, PVT not only can be trained on dense partitions of an image to achieve high output resolution, which is important for dense prediction, but also uses a progressive shrinking pyramid to reduce the computations of large feature maps. (2) PVT inherits the advantages of both CNN and Transformer, making it a unified backbone for various vision tasks without convolutions, where it can be used as a direct replacement for CNN backbones. (3) We validate PVT through extensive experiments, showing that it boosts the performance of many downstream tasks, including object detection, instance and semantic segmentation. For example, with a comparable number of parameters, PVT+RetinaNet achieves 40.4 AP on the COCO dataset, surpassing ResNet50+RetinNet (36.3 AP) by 4.1 absolute AP (see Figure 2). We hope that PVT could, serre as an alternative and useful backbone for pixel-level predictions and facilitate future research.

In this paper we present the Women in Computer Vision Workshop - WiCV 2019, organized in conjunction with CVPR 2019. This event is meant for increasing the visibility and inclusion of women researchers in the computer vision field. Computer vision and machine learning have made incredible progress over the past years, but the number of female researchers is still low both in academia and in industry. WiCV is organized especially for the following reason: to raise visibility of female researchers, to increase collaborations between them, and to provide mentorship to female junior researchers in the field. In this paper, we present a report of trends over the past years, along with a summary of statistics regarding presenters, attendees, and sponsorship for the current workshop.

Insects are the most important global pollinator of crops and play a key role in maintaining the sustainability of natural ecosystems. Insect pollination monitoring and management are therefore essential for improving crop production and food security. Computer vision facilitated pollinator monitoring can intensify data collection over what is feasible using manual approaches. The new data it generates may provide a detailed understanding of insect distributions and facilitate fine-grained analysis sufficient to predict their pollination efficacy and underpin precision pollination. Current computer vision facilitated insect tracking in complex outdoor environments is restricted in spatial coverage and often constrained to a single insect species. This limits its relevance to agriculture. Therefore, in this article we introduce a novel system to facilitate markerless data capture for insect counting, insect motion tracking, behaviour analysis and pollination prediction across large agricultural areas. Our system is comprised of edge computing multi-point video recording, offline automated multispecies insect counting, tracking and behavioural analysis. We implement and test our system on a commercial berry farm to demonstrate its capabilities. Our system successfully tracked four insect varieties, at nine monitoring stations within polytunnels, obtaining an F-score above 0.8 for each variety. The system enabled calculation of key metrics to assess the relative pollination impact of each insect variety. With this technological advancement, detailed, ongoing data collection for precision pollination becomes achievable. This is important to inform growers and apiarists managing crop pollination, as it allows data-driven decisions to be made to improve food production and food security.

Vision transformers have shown great success on numerous computer vision tasks. However, its central component, softmax attention, prohibits vision transformers from scaling up to high-resolution images, due to both the computational complexity and memory footprint being quadratic. Although linear attention was introduced in natural language processing (NLP) tasks to mitigate a similar issue, directly applying existing linear attention to vision transformers may not lead to satisfactory results. We investigate this problem and find that computer vision tasks focus more on local information compared with NLP tasks. Based on this observation, we present a Vicinity Attention that introduces a locality bias to vision transformers with linear complexity. Specifically, for each image patch, we adjust its attention weight based on its 2D Manhattan distance measured by its neighbouring patches. In this case, the neighbouring patches will receive stronger attention than far-away patches. Moreover, since our Vicinity Attention requires the token length to be much larger than the feature dimension to show its efficiency advantages, we further propose a new Vicinity Vision Transformer (VVT) structure to reduce the feature dimension without degenerating the accuracy. We perform extensive experiments on the CIFAR100, ImageNet1K, and ADE20K datasets to validate the effectiveness of our method. Our method has a slower growth rate of GFlops than previous transformer-based and convolution-based networks when the input resolution increases. In particular, our approach achieves state-of-the-art image classification accuracy with 50% fewer parameters than previous methods.

We devise a universal adaptive neural layer to "learn" optimal frequency filter for each image together with the weights of the base neural network that performs some computer vision task. The proposed approach takes the source image in the spatial domain, automatically selects the best frequencies from the frequency domain, and transmits the inverse-transform image to the main neural network. Remarkably, such a simple add-on layer dramatically improves the performance of the main network regardless of its design. We observe that the light networks gain a noticeable boost in the performance metrics; whereas, the training of the heavy ones converges faster when our adaptive layer is allowed to "learn" alongside the main architecture. We validate the idea in four classical computer vision tasks: classification, segmentation, denoising, and erasing, considering popular natural and medical data benchmarks.

Northern Bangladesh is home to the most lychee growing, which has a major economic impact. Lychee output and quality are reduced by various leaf and fruit diseases. Deep learning is used to construct a disease detection system for lychee diseases to detect them early and accurately. We correctly identified 6,000 photos of healthy and unhealthy lychee foliage and fruits. Six deep learning algorithms—VGG16, CustomCNN, MobileNet, InceptionV3, ResNet50, and Vision Transformer— classified the photos. Normalization and augmentation improved model resilience during data preparation. F1-score, recall, accuracy, and precision were utilized to train and evaluate the models. Vision Transformer (ViT) scores 99.91% in accuracy, recall, and F1-score, outperforming the other models. This achievement shows that ViT can better recognize complex lychee disease traits. The ViT algorithm for disease detection can help Bangladeshi farmers identify illnesses quickly and accurately. Reduced lychee loss and improved fruit quality and attractiveness should boost lychee sales in domestic and international markets.

With the expansion of the installed capacity of power system and the scale of power grid, the operation of various electrical equipment requires the reliability of power system to be improved continuously. Transformer is one of the key equipments in the power distribution system, and its running state is directly related to the reliability of the whole power system. Condition-based maintenance technology is to diagnose the equipment by using maintenance technology when the equipment is charged. On-line detection and fault detection technology of transformer can improve the reliability of equipment, reduce the maintenance cost of equipment and improve economic benefits. In this paper, a transformer fault detection algorithm based on computer vision and deep learning technology is proposed, and the effectiveness and feasibility of transformer computer vision fault online detection technology are verified from two aspects: fault detection scheme design and transformer fault detection system. The simulation results show that with the increase of the number of experiments, the accuracy of this algorithm is stable at about 95%, and the real-time wavelength tends to be stable, which is 15.85% higher than the traditional algorithm on average. This method can effectively detect the transformer fault, and at the same time, it can detect the latent fault that leads to the drift of transformer winding parameters.

Computer Vision Syndrome (CVS) has emerged as a critical health concern in the digital era due to prolonged exposure to screens, resulting in eye strain, dryness, and blurred vision.Traditional diagnostic methods rely on clinical examination, which can be invasive, time-consuming, and inaccessible for frequent monitoring.This paper introduces Eyelume, a real-time CVS detection and monitoring system that leverages Vision Transformers (ViT) for accurate pupil segmentation and pupillometry analysis.Unlike Convolutional Neural Networks (CNNs), ViTs capture global dependencies within visual data, offering robustness against low-quality and noisy eye images.The system enables users to upload eye images via a web interface, validates input quality, and computes pupil size variations to identify abnormal responses linked to CVS.Experimental evaluations demonstrate a segmentation accuracy of 99.6%, proving Eyelume's potential as a non-invasive, accessible, and effective tool for early CVS detection and digital eye health monitoring [1][2].

Vision Transformer (ViT) is an image recognition model that uses transformer architecture, which has a numerous advantage over Convolution Neural Networks (CNN). It offers improved accuracy, scalability, flexibility, global context, and transferability. ViT can handle images of different sizes and aspect ratios, making it more versatile than CNN. It can process an entire image at once, allowing it to capture global context information and long-range dependencies. Additionally, ViTs pre-training on huge amounts of image data can be transferred to other image recognition tasks, making it a useful tool for transfer learning. This paper describes the differences between ViT and CNN and how ViT splits images into patches for classification. The positional encoding of different features is done in ViT to avoid the requirement of filters. Proposed implementation obtained final accuracy of prediction 93% for top-1 accuracy.