Mahdi Rezaei

University of Leeds · Institute for Transport Studies (ITS)

The challenge of learning a new concept without receiving any examples beforehand is called zero-shot learning (ZSL). One of the major issues in deep learning based methodologies is the requirement of feeding a vast amount of annotated and labelled images by a human to train the network model. ZSL is known for having minimal human intervention by r...

Scene understanding plays a crucial role in autonomous driving by utilizing sensory data for contextual information extraction and decision making. Beyond modeling advances, the enabler for vehicles to become aware of their surroundings is the availability of visual sensory data, which expand the vehicular perception and realizes vehicular contextu...

Following the successful development of Advanced Driver Assistance Systems (ADAS), the current research directions focus on higher levels of driving automation aiming at reducing human driving tasks and extending the operational design domain (ODD), while maintaining a higher level of safety in automated driving mode. Currently, there are high dema...

Machine Learning has played a major role in various applications including Autonomous Vehicles and Intelligent Transportation Systems. Utilizing a deep convolutional neural network, the article introduces a zero-calibration 3D Object recognition and tracking system for traffic monitoring. The model can accurately work on urban traffic cameras, rega...

Social distancing is a recommended solution by the World Health Organisation (WHO) to minimise the spread of COVID-19 in public places. The majority of governments and national health authorities have set the 2-m physical distancing as a mandatory safety measure in shopping centres, schools and other covered areas. In this research, we develop a hy...

Recent advancements in predicting pedestrian crossing intentions for Autonomous Vehicles using Computer Vision and Deep Neural Networks are promising. However, the black-box nature of DNNs poses challenges in understanding how the model works and how input features contribute to final predictions. This lack of interpretability delimits the trust in...

Adverse conditions like snow, rain, nighttime, and fog, pose challenges for autonomous driving perception systems. Existing methods have limited effectiveness in improving essential computer vision tasks, such as semantic segmentation, and often focus on only one specific condition, such as removing rain or translating nighttime images into daytime...

Multi-task learning in advanced driver assistance systems aims to endow models with the capacity to jointly handle multiple related tasks, such as object detection, depth estimation, and more. However, existing multi-task learning models largely rely on the extensive number of labelled data. In practice, the process of annotating data for multi-tas...

Autonomous vehicles (AVs) are becoming an indispensable part of future transportation. However, safety challenges and lack of reliability limit their real-world deployment. Towards boosting the appearance of AVs on the roads, the interaction of AVs with pedestrians including “prediction of the pedestrian crossing intention” deserves extensive resea...

The development of Adaptive Cruise Control (ACC) systems aims to enhance the safety and comfort of vehicles by automatically regulating the speed of the vehicle to ensure a safe gap from the preceding vehicle. However, conventional ACC systems are unable to adapt themselves to changing driving conditions and drivers' behavior. To address this limit...

Accurate lane change prediction can reduce potential accidents and contribute to higher road safety. Adaptive cruise control (ACC), lane departure avoidance (LDA), and lane keeping assistance (LKA) are some conventional modules in advanced driver assistance systems (ADAS). Thanks to vehicle-to-vehicle communication (V2V), vehicles can share traffic...

Autonomous vehicles (AVs) are becoming an indispensable part of future transportation. However, safety challenges and lack of reliability limit their real-world deployment. Towards boosting the appearance of AVs on the roads, the interaction of AVs with pedestrians including "prediction of the pedestrian crossing intention" deserves extensive resea...

The acceptance of AI-based intelligent transportation systems requires addressing the existing barriers and the adoption of macro-decisions and policies by policymakers and governments. This study evaluates the potential barriers to the adoption of Autonomous Vehicles (AVs) in developing countries by considering the sustainability dimensions. The b...

Aiming at highly accurate object detection for connected and automated vehicles (CAVs), this paper presents a Deep Neural Network based 3D object detection model that leverages a three-stage feature extractor by developing a novel LIDAR-Camera fusion scheme. The proposed feature extractor extracts high-level features from two input sensory modaliti...

Computer Vision has played a major role in Intelligent Transportation Systems (ITS) and traffic surveillance. Along with the rapidly growing automated vehicles and crowded cities, the automated and advanced traffic management systems (ATMS) using video surveillance infrastructures have been evolved by the implementation of Deep Neural Networks. In...

Aiming at highly accurate object detection for connected and automated vehicles (CAVs), this paper presents a Deep Neural Network based 3D object detection model that leverages a three-stage feature extractor by developing a novel LIDAR-Camera fusion scheme. The proposed feature extractor extracts high-level features from two input sensory modaliti...

One of the most challenging and non-trivial tasks in robot-based rescue operations is the Hazardous Materials (HAZMAT) sign detection in dangerous operation fields, in order to prevent further unexpected disasters. Each HAZMAT sign has a specific meaning that the rescue robot should detect and interpret it to take a safe action, accordingly. Accura...

Reward engineering and designing an incentive reward function are non-trivial tasks to train agents in complex environments. Furthermore, an inaccurate reward function may lead to a biased behaviour which is far from an efficient and optimised behaviour. In this paper, we focus on training a single agent to score goals with binary success/failure r...

Automated human action recognition is one of the most attractive and practical research fields in computer vision. In such systems, the human action labelling is based on the appearance and patterns of the motions in the video sequences; however, majority of the existing research and most of the conventional methodologies and classic neural network...

In this paper, we present an active vision method using a deep reinforcement learning approach for a humanoid soccer-playing robot. The proposed method adaptively optimises the viewpoint of the robot to acquire the most useful landmarks for self-localisation while keeping the ball into its viewpoint. Active vision is critical for humanoid decision-...

In this paper, we present an active vision method using a deep reinforcement learning approach for a humanoid soccer-playing robot. The proposed method adaptively optimises the viewpoint of the robot to acquire the most useful landmarks for self-localisation while keeping the ball into its viewpoint. Active vision is critical for humanoid decision-...

Social distancing is a recommended solution by the World Health Organisation (WHO) to minimise the spread of COVID-19 in public places. The majority of governments and national health authorities have set the 2-meter physical distancing as a mandatory safety measure in shopping centres, schools and other covered areas. In this research, we develop...

The challenge of learning a new concept, object, or a new medical disease recognition without receiving any examples beforehand is called Zero-Shot Learning (ZSL). One of the major issues in deep learning based methodologies such as in Medical Imaging and other real-world applications is the requirement of large annotated datasets prepared by clini...

Social distancing is a recommended solution by the World Health Organisation (WHO) to minimise the spread of COVID-19 in public places. The majority of governments and national health authorities have set the 2-meter physical distancing as a mandatory safety measure in shopping centres, schools and other covered areas. In this research, we develop...

One of the most challenging and non-trivial tasks in robotics-based rescue operations is Hazardous Materials or HAZMATs sign detection within the operation field, in order to prevent other unexpected disasters. Each Hazmat sign has a specific meaning that the rescue robot should detect and interpret it to take a safe action, accordingly. Accurate H...

Automated human action recognition is one of the most attractive and practical research fields in computer vision, in spite of its high computational costs. In such systems, the human action labelling is based on the appearance and patterns of the motions in the video sequences; however, the conventional methodologies and classic neural networks ca...

Automated human action recognition is one of the most attractive and practical research fields in computer vision, in spite of its high computational costs. In such systems, the human action labelling is based on the appearance and patterns of the motions in the video sequences; however, the conventional methodologies and classic neural networks ca...

CAPTCHA is a human-centred test to distinguish a human operator from bots, attacking programs, or any other computerised agent that tries to imitate human intelligence. In this research, we investigate a way to crack visual CAPTCHA tests by an automated deep learning based solution. The goal of the cracking is to investigate the weaknesses and vuln...

Automated human action recognition is one of the most attractive and practical research felds in computer vision. In such systems, the human action labelling is based on the appearance and patterns of the motions in the video sequences; however, majority of the existing research and most of the conventional methodologies and classic neural networ...

Ball detection is one of the most important tasks in the context of soccer-playing robots. The ball is a small moving object which can be blurred and occluded in many situations. Several neural network based methods with different architectures are proposed to deal with the ball detection. However, they are either neglecting to consider the computa...

Ball detection is one of the most important tasks in the context of soccer-playing robots. The ball is a small moving object which can be blurred and occluded in many situations. Several neural network based methods with different architectures are proposed to deal with the ball detection. However, they are either neglecting to consider the comput...

This paper highlights the role of ground manifold modeling for stixel calculations; stixels are medium-level data representations used for the development of computer vision modules for self-driving cars. By using single-disparity maps and simplifying ground manifold models, calculated stixels may suffer from noise, inconsistency, and false-detecti...

Accurate license plate localization is the most important prerequisite in ANPR (Automatic Number Plate Recognition) systems. Majority of the existing algorithms use a single feature to obtain the license plate location which causes to potential false detections. In this article we propose a robust methodology using 16 statistical features while we...

As part of the published research in [1], [2] titled Mahdi Rezaei, Mutsuhiro Terauchi, "Vehicle Detection Based on Multi-Feature Clues and Dempster-Shafer Fusion Theory", Image and Video Technology, Springer, Volume 8333, pp 60-72, 2014, we release the iROADS as a comprehensive vehicle-dataset with 4656 image frames, in 7 categories, recorded from...

Stixel-based segmentation is specifically designed towards obstacle detection which combines road surface estimation in traffic scenes, stixel calculations, and stixel clustering. Stixels are defined by observed height above road surface. Road surfaces (ground manifolds) are represented by using an occupancy grid map. Stixel-based segmentation may...

We present a novel method for stixel construction using a calibrated collinear trinocular vision system. Our method takes three conjugate stereo images at the same time to measure the consistency of disparity values by means of the transitivity error in disparity space. Unlike previous stixel estimation methods that are built based on a single disp...

Face appearance is one of the most important visual features of human which varies significantly over the aging. Therefore, automatic age estimation is a demanding research topic in the field of facial feature analysis. In the task of age estimation, feature extraction is the first influential step which highly effects on a learning method and its...

This chapter mainly proposes a comprehensive method for detecting driver’s distraction and inattention. We introduce an asymmetric appearance-modelling method and an accurate 2D-to-3D registration technique to obtain the driver’s head pose, yawing detection, and head-nodding detection. Chapter 5 and this chapter present the first major objective of...

In this chapter we present and discuss the basic computer vision concepts, techniques, and mathematical background that we use in this book. The chapter introduces image notations, the concept of integral images, colour space conversions, the Hough transform for line detection, camera coordinate systems, and stereo computer vision.

This chapter outlines the general context of the book. Autonomous driving is still at a stage where drivers are expected to be in control of the vehicle at all times, but provided automated control features of the vehicle (based on input data generated by different sensors) already enhance safety and driver comfort. We especially consider automated...

This book focuses in particular on driver-environment understanding as briefly outlined at the end of the previous chapter. This chapter provides a more detailed introduction, motivations, and a review of the state-of-the-art in this area of vision-based driver-assistance systems. The chapter also discusses existing challenges and outlines the stru...

In this chapter we propose a method to assess driver drowsiness based on face and eye-status analysis. The chapter starts with a detailed discussion on effective ways to create a strong classifier (the “training phase”), and it continues with a novel optimization method for the “application phase” of the classifier. Both together significantly impr...

In this chapter we discuss how to assess the risk level in a given driving scenario based on the eight possible inputs: driver’s direction of attention (yaw, roll, pitch), signs of fatigue or drowsiness (yawning, head nodding, eye closure), and from road situations (distance, and the angle of the detected vehicles to the ego-vehicle). Using a fuzzy...

In this chapter we outline object detection and object recognition techniques which are of relevance for the remainder of the book. We focus on supervised and unsupervised learning approaches. The chapter provides technical details for each method, discussions on the strengths and weaknesses of each method, and gives examples and various applicatio...

“Collision warning systems” are actively researched in the area of computer vision and the automotive industry. Using monocular vision only, this chapter discusses the part of our study that aims at detecting and tracking the vehicles ahead, to identify safety distances, and to provide timely information to assist a distracted driver under various...

https://www.amazon.com/Computer-Vision-Driver-Assistance-Computational/dp/3319505491/ref=sr_1_1?s=books&ie=UTF8&qid=1488803742&sr=1-1

Computer-vision based driver assistance is an emerging technology, in both automotive industry and academia. Despite the existence of some commercial safety systems such as night vision, adaptive cruise control, and lane departure warning systems, we are at the beginning of a long research pathway toward future generation of intelligent vehicles. C...

Avoiding high computational costs and calibration issues involved in stereo-vision based algorithms, this article proposes real-time monocular-vision based techniques for simultaneous vehicle detection and inter-vehicle distance estimation, in which the performance and robustness of the system remain competitive, even for highly challenging benchma...

Fatigue and driver drowsiness monitoring is an important subject for designing driver assistance systems. The measurement of eye closure is a fundamental step for driver awareness detection. We propose a method which is based on eyelid detection and the measurement of the distance between the eyelids. First, the face and the eyes of the driver are...

Eye gaze detection under challenging lighting conditions is a non-trivial task. Pixel intensity and the shades around the eye region may change depending on the time of day, location, or due to artificial lighting. This paper introduces a lighting-adaptive solution for robust eye gaze detection. First, we propose a binarization and cropping techniq...

Artistic rendering of human portraits is different and more challenging than that of landscapes or flowers. Issues are eye, nose, and mouth regions (i.e., facial features) where we need to represent their natural emotions. Shades or darkness around eyes, or shininess at nose tips may negatively impact the rendering result if not properly dealt with...

[Best PhD Student Published Paper Award 2014] The paper proposes an advanced driver-assistance system that correlates the driver’s attention to the road and traffic conditions by analyzing both simultaneously. In particular, we aim at the prevention of rear-end crashes due to driver fatigue or distraction. We propose an asymmetric appearance-modeli...

Please refer to the updated iROADS dataset with the added Ground Truth information via the following link: https://www.researchgate.net/publication/260219141_iROADS_Datasetzip

Please refer to the updated iROADS dataset with the added Ground Truth information via the following link: https://www.researchgate.net/publication/260219141_iROADS_Datasetzip

On-road vehicle detection and rear-end crash prevention are demanding subjects in both academia and automotive industry. The paper focuses on monocular vision-based vehicle detection under challenging lighting conditions, being still an open topic in the area of driver assistance systems. The paper proposes an effective vehicle detection method bas...

This paper addresses the problem of detecting human faces in noisy images. We propose a method that includes a denoising preprocessing step, and a new face detection approach based on a novel extension of Haar-like features. Preprocessing of the input images is focused on the removal of different types of noise while preserving the phase data. For...

The paper develops a novel technique that significantly improves the performance of Haar-like feature-based object detectors in terms of speed, detection rate under difficult lighting conditions, and reduced number of false-positives. The method is implemented and validated for driver monitoring under very dark, very bright, and normal conditions....

The paper introduces a novel methodology to enhance the accuracy, performance and effectiveness of Haar-like classifiers, especially for complicated lighting conditions. Performing a statistical intensity analysis on input image sequences, the technique provides a very fast and robust eye-status detection via a low-resolution VGA camera, without ap...

Artistic painting of human portraits are more challenging than landscapes or flowers. Challenges are eye and nose areas where we need to avoid alterations from their natural appearance. Shades or darkness around eyes, or shininess at the nose tip may negatively impact the rendering result if not properly dealt with. The proposed computerized method...

The design of intelligent driver assistance systems is of increasing importance for the vehicle-producing industry and road-safety solutions. This article starts with a review of road-situation monitoring and driver's behaviour analysis. This article also discusses lane tracking using vision (or other) sensors, and the strength or weakness of diffe...

Eye status detection and localization is a fundamental step for driver awareness detection. The efficiency of any learning-based object detection method highly depends on the training dataset as well as learning parameters. The research develops optimum values of Haar-training parameters to create a nested cascade of classifiers for real-time eye s...

This article contributes to a novel computerized and heuristic algorithm to make pretty geometrics used in traditional building patterns and tiling. Starting with some basic sketch or an ornamental pattern, it depicts how a designer may analyze the sketch structure and approaches to an analytical representation. In return, this analytical represent...

This article contributes to a novel computerized and heuristic algorithm to make pretty geometrics used in traditional building patterns and tiling. Starting with some basic sketch or an ornamental pattern, it depicts how a designer may analyze the sketch structure and approaches to an analytical representation. In return, this analytical represent...