A5057980050- Human Pose and Action Recognition
- Advanced Image and Video Retrieval Techniques
- Video Surveillance and Tracking Methods
- Image Retrieval and Classification Techniques
- Advanced Neural Network Applications
- Context-Aware Activity Recognition Systems
- Anomaly Detection Techniques and Applications
- Video Analysis and Summarization
- Domain Adaptation and Few-Shot Learning
- Visual Attention and Saliency Detection
- Handwritten Text Recognition Techniques
- Data Visualization and Analytics
- Autonomous Vehicle Technology and Safety
- Digital Media Forensic Detection
- Hand Gesture Recognition Systems
- Face recognition and analysis
- Image Enhancement Techniques
- Music and Audio Processing
- Retinal Imaging and Analysis
- Hydrological Forecasting Using AI
- Advanced Vision and Imaging
- Energy Load and Power Forecasting
- Human-Automation Interaction and Safety
- Retinal Diseases and Treatments
- Teleoperation and Haptic Systems
Edge Hill University
2015-2025
University of Leeds
2010-2015
University of Fribourg
2004-2007
Abstract Non-predictive or inaccurate weather forecasting can severely impact the community of users such as farmers. Numerical prediction models run in major centers with several supercomputers to solve simultaneous complex nonlinear mathematical equations. Such provide medium-range forecasts, i.e., every 6 h up 18 grid length 10–20 km. However, farmers often depend on more detailed short-to forecasts higher-resolution regional models. Therefore, this research aims address by developing and...
Abstract It is well-known that numerical weather prediction (NWP) models require considerable computer power to solve complex mathematical equations obtain a forecast based on current conditions. In this article, we propose novel lightweight data-driven forecasting model by exploring temporal modelling approaches of long short-term memory (LSTM) and convolutional networks (TCN) compare its performance with the existing classical machine learning approaches, statistical dynamic ensemble...
Deep convolutional neural networks (CNNs) have shown a strong ability in mining discriminative object pose and parts information for image recognition. For fine-grained recognition, context-aware rich feature representation of object/scene plays key role since it exhibits significant variance the same subcategory subtle among different subcategories. Finding that fully characterizes is not straightforward. To address this, we propose novel attentional pooling (CAP) effectively captures...
Over the past few years, a significant progress has been made in deep convolutional neural networks (CNNs)-based image recognition. This is mainly due to strong ability of such mining discriminative object pose and parts information from texture shape. often inappropriate for fine-grained visual classification (FGVC) since it exhibits high intra-class low inter-class variances occlusions, deformation, illuminations, etc. Thus, an expressive feature representation describing global structural...
Background Optical coherence tomography angiography (OCTA) enables fast and non-invasive high-resolution imaging of retinal microvasculature is suggested as a potential tool in the early detection microvascular changes Alzheimer’s Disease (AD). We developed standardised OCTA analysis framework compared their extracted parameters among controls AD/mild cognitive impairment (MCI) cross-section study. Methods defined geometrical at different layers foveal avascular zone (FAZ) from segmented...
Abstract Learning involves a substantial amount of cognitive, social and emotional states. Therefore, recognizing understanding these states in the context learning is key designing informed interventions addressing needs individual student to provide personalized education. In this paper, we explore automatic detection learner’s nonverbal behaviors involving hand-over-face gestures, head eye movements emotions via facial expressions during learning. The proposed computer vision-based...
Automated detection of retinal structures, such as vessels (RV), the foveal avascular zone (FAZ), and vascular junctions (RVJ), are great importance for understanding diseases eye clinical decision-making. In this paper, we propose a novel Voting-based Adaptive Feature Fusion multi-task network (VAFF-Net) joint segmentation, detection, classification RV, FAZ, RVJ in optical coherence tomography angiography (OCTA). A task-specific voting gate module is proposed to adaptively extract fuse...
Today, the workflows that are involved in industrial assembly and production activities becoming increasingly complex. To efficiently safely perform these is demanding on workers, particular when it comes to infrequent or repetitive tasks. This burden workers can be eased by introducing smart assistance systems. article presents a scalable concept an integrated system demonstrator designed for this purpose. The basic idea learn from observing multiple expert operators then transfer learnt...
Background: Medical robots are increasingly used for a variety of applications in healthcare. Robots have mainly been to support surgical procedures, and assistive uses dementia elderly care. To date, there has limited debate about the potential opportunities risks robotics other areas palliative, supportive end-of-life Aim: The objective this article is examine possible future impact medical on care Specifically, we will discuss strengths, weaknesses, threats (SWOT) technology. Methods: A...
Automatic recognition and prediction of in-vehicle human activities has a significant impact on the next generation driver assistance intelligent autonomous vehicles. In this article, we present novel single image action algorithm inspired by perception that often focuses selectively parts images to acquire information at specific places which are distinct given task. Unlike existing approaches, argue activity is combination pose semantic contextual cues. detail, model considering...
Autonomous vehicles (AVs) are undergoing rapid worldwide development. They will only become a success if they accepted by their users. Therefore, there is need for user acceptance these vehicles. Previous studies on of AV have identified several predictors. Inspired studies, the authors’ investigation aimed at sociodemographic characteristics, including broader individual and societal acceptance, beyond technical issues to get clear view acceptance. In this study, surveyed 229 respondents,...
Accurate detection of objects from LiDAR point clouds is crucial for autonomous driving and environment modeling. However, uncertainties in ground truth labels due to occlusions, sparsity, truncation can hinder model training performance. This paper introduces two strategies address these issues: 1) Soft Regression Loss (SoRL) 2) Discrete Quantization Sampling (DQS). SoRL utilizes Gaussian distributions object predictions, measuring uncertainty based on the probability within distributions....
The following topics are dealt with: object detection; learning (artificial intelligence); feature extraction; video surveillance; signal processing; image classification; convolutional neural nets; motion analysis; computer vision; tracking.
There is significant progress in recognizing traditional human activities from videos focusing on highly distinctive actions involving discriminative body movements, body-object and/or human-human interactions. Driver's are different since they executed by the same subject with similar parts resulting subtle changes. To address this, we propose a novel framework exploiting spatiotemporal attention to model Our named Coarse Temporal Attention Network (CTA-Net), which coarse temporal branches...
Abstract This paper presents a novel approach for Fine-Grained Visual Classification (FGVC) by exploring Graph Neural Networks (GNNs) to facilitate high-order feature interactions, with specific focus on constructing both inter- and intra-region graphs. Unlike previous FGVC techniques that often isolate global local features, our method combines features seamlessly during learning via Inter-region graphs capture long-range dependencies recognize patterns, while delve into finer details...
Perception of scenes has typically been investigated by using static or simplified visual displays. How attention is used to perceive and evaluate dynamic, realistic more poorly understood, in part due the problem comparing eye fixations moving stimuli across observers. When task stimulus common observers, consistent fixation location can indicate that region high goal-based relevance. Here we these issues when an observer a specific, naturalistic, task: closed-circuit television (CCTV)...
Human pose estimation through deep learning has achieved very high accuracy over various difficult poses. However, these are computationally expensive and often not suitable for mobile based systems. In this paper, we investigate the use of MobileNets, which is well-known to be a light-weight efficient CNN architecture embedded vision applications. We adapt MobileNets inspired by hourglass network. introduce novel split stream at final two layers MobileNets. This approach reduces...