Tania Stathaki

• PhD
• Professor (Associate) at Imperial College London

The growth in the use of sensor technology has led to the demand for image fusion: signal processing techniques that can combine information received from different sensors into a single composite image in an efficient and reliable manner. This book brings together classical and modern algorithms and design architectures, demonstrating through appl...

The task of enhancing the perception of a scene by combining information captured by different sensors is usually known as image fusion. The pyramid decomposition and the Dual-Tree Wavelet Transform have been thoroughly applied in image fusion as analysis and synthesis tools. Using a number of pixel-based and region-based fusion rules, one can comb...

We present a robust FFT-based approach to scale-invariant image registration. Our method relies on FFT-based correlation twice: once in the log-polar Fourier domain to estimate the scaling and rotation and once in the spatial domain to recover the residual translation. Previous methods based on the same principles are not robust. To equip our schem...

We propose a robust approach to discriminant kernel-based feature extraction for face recognition and verification. We show, for the first time, how to perform the eigen analysis of the within-class scatter matrix directly in the feature space. This eigen analysis provides the eigenspectrum of its range space and the corresponding eigenvectors as w...

A normalized robust mixed-norm (NRMN) algorithm for system identification in the presence of impulsive noise is introduced. The standard robust mixed-norm (RMN) algorithm exhibits slow convergence, requires a stationary operating environment, and employs a constant step-size that needs to be determined a priori. To overcome these limitations, the p...

This paper investigates the impact of various data augmentation techniques on the performance of object detection models. Specifically, we explore classical augmentation methods, image compositing, and advanced generative models such as Stable Diffusion XL and ControlNet. The objective of this work is to enhance model robustness and improve detecti...

Contrastive learning has become a dominant approach in self-supervised visual representation learning, with hard negatives-samples that closely resemble the anchor-being key to enhancing the discriminative power of learned representations. However, efficiently leveraging hard negatives remains a challenge due to the difficulty in identifying and in...

Scribble supervision, a common form of weakly supervised learning, involves annotating pixels using hand-drawn curve lines, which helps reduce the cost of manual labelling. This technique has been widely used in medical image segmentation tasks to fasten network training. However, scribble supervision has limitations in terms of annotation consiste...

The design of pedestrian detectors seldom considers the unique characteristics of this task and usually follows the common strategies for general object detection. To explore the potential of these characteristics, we take the perspective effect in pedestrian datasets as an example and propose the mean height aided suppression for post-processing....

Knowledge distillation (KD) involves transferring the knowledge from one neural network to another, often from a larger, well-trained model (teacher) to a smaller, more efficient model (student). Traditional KD methods minimize the Kullback-Leibler (KL) divergence between the probabilistic outputs of the teacher and student networks. However, this...

Knowledge distillation (KD) is an effective method for transferring knowledge from a large, well-trained teacher model to a smaller, more efficient student model. Despite its success, one of the main challenges in KD is ensuring the efficient transfer of complex knowledge while maintaining the student's computational efficiency. Unlike previous wor...

Transductive few-shot learning algorithms have showed substantially superior performance over their inductive counterparts by leveraging the unlabeled queries. However, the vast majority of such methods are evaluated on perfectly class-balanced benchmarks. It has been shown that they undergo remarkable drop in performance under a more realistic, im...

In the current context of climate change and demographic expansion, one of the phenomena that humanity faces are the suburban wildfires. To prevent the occurrence of suburban forest fires, fire risk assessment and early fire detection approaches need to be applied. Forest fire risk mapping depends on various factors and contributes to the identific...

Gastric cancer is one of the most frequent causes of cancer-related deaths worldwide. Gastric atrophy (GA) and gastric intestinal metaplasia (IM) of the mucosa of the stomach have been found to increase the risk of gastric cancer and are considered precancerous lesions. Therefore, the early detection of GA and IM may have a valuable role in histopa...

Vision transformers have demonstrated remarkable performance on a variety of computer vision tasks. In this paper, we illustrate the effectiveness of the deformable vision transformer for single-stage pedestrian detection and propose a spatial and multi-scale feature enhancement module, which aims to achieve the optimal balance between speed and ac...

Polyp segmentation is a crucial step towards computer-aided diagnosis of colorectal cancer. However, most of the polyp segmentation methods require pixel-wise annotated datasets. Annotated datasets are tedious and time-consuming to produce, especially for physicians who must dedicate their time to their patients. We tackle this issue by proposing a...

Gastric cancer is one of the most common cancers and a leading cause of cancer-related death worldwide. Among the risk factors of gastric cancer, the gastric intestinal metaplasia (IM) has been found to increase the risk of gastric cancer and is considered as one of the precancerous lesions. Therefore, early detection of IM could allow risk stratif...

RGB-D salient object detection (SOD) demonstrates its superiority in detecting in complex environments due to the additional depth information introduced in the data. Inevitably, an independent stream is introduced to extract features from depth images, leading to extra computation and parameters. This methodology sacrifices the model size to impro...

Salient object detection (SOD) aims at locating the most significant object within a given image. In recent years, great progress has been made in applying SOD on many vision tasks. The depth map could provide additional spatial prior and boundary cues to boost the performance. Combining the depth information with image data obtained from standard...

Pedestrian detection is a challenging task, mainly owing to the numerous appearances of human bodies. Modern detectors extract representative features via the deep neural network; however, they usually require a large training set and high-performance GPUs. For these cases, we propose a novel human detection approach that integrates a pretrained fa...

Over the last decade or so, laser scanning technology has become an increasingly popular and important tool for forestry inventory, enabling accurate capture of 3D information in a fast and environmentally friendly manner. To this end, the authors propose here a system for tropical tree species classification based on 3D scans of LiDAR sensing tech...

Nowadays the civil infrastructure is exposed to several challenges such as daily vehicular traffic and extreme weather conditions, i.e. ghastly winds, strong rain etc. It is well known that these may determine structural deterioration and damages, which can even cause catastrophic collapses related to significant socio-economic losses. For this rea...

Nowadays the civil infrastructure is exposed to several challenges such as daily vehicular traffic and extreme weather conditions, i.e. ghastly winds, strong rain etc. It is well known that these may determine structural deterioration and damages, which can even cause catastrophic collapses related to significant socio-economic losses. For this rea...

We consider the problem of highway ramp metering with Model Predictive Control (MPC). While MPC is considered one of the most robust approaches for ramp metering, the optimization problem that has to be solved is often large, nonlinear, and nonconvex, making its real-time implementation prohibitive. To deal with this issue, we develop a novel Expli...

We consider the problem of highway ramp metering with Model Predictive Control (MPC). While MPC is considered one of the most robust approaches for ramp metering, the optimization problem that has to be solved is often large, nonlinear, and nonconvex, making its real-time implementation prohibitive. To deal with this issue, we develop a novel Expli...

Nowadays the civil infrastructure is exposed to several challenges such as daily vehicular traffic and extreme weather conditions, i.e. ghastly winds, strong rain etc. It is well known that these may determine structural deterioration and damages, which can even cause catastrophic collapses related to significant socioeconomic losses. For this reas...

RGB-D salient object detection(SOD) demonstrates its superiority on detecting in complex environments due to the additional depth information introduced in the data. Inevitably, an independent stream is introduced to extract features from depth images, leading to extra computation and parameters. This methodology which sacrifices the model size to...

Few-shot classification addresses the challenge of classifying examples given not just limited supervision but limited data as well. An attractive solution is synthetic data generation. However, most such methods are overly sophisticated, focusing on high-quality, realistic data in the input space. It is unclear whether adapting them to the few-sho...

Salient object detection(SOD) aims at locating the most significant object within a given image. In recent years, great progress has been made in applying SOD on many vision tasks. The depth map could provide additional spatial prior and boundary cues to boost the performance. Combining the depth information with image data obtained from standard v...

The current state-of-the-art hand gesture recognition methodologies heavily rely in the use of machine learning. However there are scenarios that machine learning cannot be applied successfully, for example in situations where data is scarce. This is the case when one-to-one matching is required between a query and a dataset of hand gestures where...

With the objective of quality assessment, cracks on concrete buildings have to be identified and monitored continuously. Due to the availability of cheap devices, techniques based on image processing have been gaining in popularity, but they require a rigorous analysis of large amounts of data. Moreover, the detection of fractures in images is stil...

Over the last decade the civil infrastructure has been experiencing a diffused deterioration that raises several concerns related to its economic and environmental impact. For this reason, extending the service life of structures has become a key objective in the field of building engineering. In particular, self-repairing materials can play an imp...

Nowadays the civil infrastructure is exposed to several challenges such as daily vehicular traffic and extreme weather conditions, i.e. ghastly winds, strong rain etc. It is well known that these may determine structural deterioration and damages, which can even cause catastrophic collapses related to significant socioeconomic losses. For this reas...

Few-shot classification addresses the challenge of classifying examples given only limited labeled data. A powerful approach is to go beyond data augmentation, towards data synthesis. However, most of data augmentation/synthesis methods for few-shot classification are overly complex and sophisticated, e.g. training a wGAN with multiple regularizers...

The current state-of-the-art hand gesture recognition methodologies heavily rely in the use of machine learning. However there are scenarios that machine learning cannot be applied successfully, for example in situations where data is scarce. This is the case when one-to-one matching is required between a query and a database of hand gestures where...

Few-shot learning amounts to learning representations and acquiring knowledge such that novel tasks may be solved with both supervision and data being limited. Improved performance is possible by transductive inference, where the entire test set is available concurrently, and semi-supervised learning, where more unlabeled data is available. These p...

Salient object detection has achieved great improvements by using the Fully Convolutional Networks (FCNs). However, the FCN-based U-shape architecture may cause dilution problems in the high-level semantic information during the up-sample operations in the top-down pathway. Thus, it can weaken the ability of salient object localization and produce...

The environmental challenges the world faces have never been greater or more complex. Global areas that are covered by forests and urban woodlands are threatened by large-scale forest fires that have increased dramatically during the last decades in Europe and worldwide, in terms of both frequency and magnitude. To this end, rapid advances in remot...

Salient object detection has achieved great improvement by using the Fully Convolution Network (FCN). However, the FCN-based U-shape architecture may cause the dilution problem in the high-level semantic information during the up-sample operations in the top-down pathway. Thus, it can weaken the ability of salient object localization and produce de...

The advent of powerful image processing and machine learning tools has generated opportunities for more automated and efficient paleography. In this work, methods were developed to automate aspects of the transcription process of ancient papyri. First, a technique is proposed that employs color thresholding and contouring for the automatic extracti...

In this paper, in order to contribute to the protection of the value and potential of forest ecosystems and global forest future we propose a novel fire detection framework, which combines recently introduced 360-degree remote sensing technology, multidimensional texture analysis and deep convolutional neural networks. Once 360-degree data are obta...

Over the last decade or so, laser scanning technology has become an increasingly popular and important tool for forestry inventory, enabling accurate capture of 3D information in a fast and environmentally friendly manner. To this end, the authors propose here a system for tropical tree species classification based on 3D scans of LiDAR sensing tech...

Pedestrian detection methods have been significantly improved with the development of deep convolutional neural networks. Nevertheless, detecting ismall-scaled pedestrians and occluded pedestrians remains a challenging problem. In this paper, we propose a pedestrian detection method with a couple-network to simultaneously address these two issues....

In recent years, the effects of climate change have become more intense and the role played by the forest ecosystems and the forest biomass produced, seems to become increasingly important. Forest biomass is a source of large quantities of timber, but also of a great variety of valuable materials, chemicals and biofuels, and therefore, the identifi...

Pedestrian detection methods have been significantly improved with the development of deep convolutional neural networks. Nevertheless, detecting small-scaled pedestrians and occluded pedestrians remains a challenging problem. In this paper, we propose a pedestrian detection method with a couple-network to simultaneously address these two issues. O...

Given the urgent priority around protecting the forests and limiting the impacts of the climate change, the constant monitoring of forests towards the achievement of accurate and timely detection of infestations and the catastrophic action of invasive insects, pests and fungi is an important and challenging task. More precisely, new species of inse...

Convolutional neural networks (CNNs) have resurged lately due to their state-of-the-art performance in various disciplines, such as computer vision, audio and text processing. However, CNNs have not been widely employed for remote sensing applications. In this paper, we propose a CNN architecture, named Modular-CNN, to improve the performance of bu...

Pedestrian detection methods have been significantly improved with the development of deep convolutional neural networks. Nevertheless, robustly detecting pedestrians with a large variant on sizes and with occlusions remains a challenging problem. In this paper, we propose a gated multi-layer convolutional feature extraction method which can adapti...

In this paper, a crack detection mechanism for concrete tunnel surfaces is presented. The proposed methodology leverages deep Convolutional Neural Networks and domain-specific heuristic post-processing techniques to address a variety of challenges, including high accuracy requirements, low operational times and limited hardware resources, poor and...

Invasive insect pests and fungi, which are introduced accidentally to forests and affect tree growth and survival, constitute a serious threat for the forests and trees acting on climate change and its impacts. Thus, the need for early and accurate health determination process of forest regions, has significantly increased the interest in automatic...

Given the broad range of applications from video surveillance to human–computer interaction, human action learning and recognition analysis based on 3D skeleton data are currently a popular area of research. In this paper, we propose a method for action recognition using depth sensors and representing the skeleton time series sequences as higher-or...

Phase correlation (PC) is widely employed by several sub-pixel motion estimation techniques in an attempt to accurately and robustly detect the displacement between two images. To achieve sub-pixel accuracy, these techniques employ interpolation methods and function-fitting approaches on the cross-correlation function derived from the PC core. Howe...

Convolutional neural networks (CNN) have enabled significant improvements in pedestrian detection owing to the strong representation ability of the CNN features. However, it is generally difficult to reduce false positives on hard negative samples such as tree leaves, traffic lights, poles, etc. Some of these hard negatives can be removed by making...

Convolutional neural networks (CNN) have enabled significant improvements in pedestrian detection owing to the strong representation ability of the CNN features. Recently, aggregating features from multiple layers of a CNN has been considered as an effective approach, however, the same approach regarding feature representation is used for detecting...

In this paper, a thorough theoretical analysis on the construction of multi-dimensional directional steerable filters is given. Steerable filters have been constructed for up to three dimensions. We extend the relevant theory to multiple dimensions and construct multi-dimensional steerable filters, as well as quadrature pairs of such filters. Formu...

In this paper, the problem of joint disparity and motion estimation from stereo image sequences is formulated in the spatiotemporal frequency domain, and a novel steerable filter-based approach is proposed. Our rationale behind coupling the two problems is that according to experimental evidence in the literature, the biological visual mechanisms f...

A strategy is introduced for achieving high accuracy in synthetic aperture radar (SAR) automatic target recognition (ATR) tasks. Initially, a novel pose rectification process and an image normalization process are sequentially introduced to produce images with less variations prior to the feature processing stage. Then, feature sets that have a wea...

With the ever-increasing demand in the analysis and understanding of aerial images in order to remotely recognise targets, this paper introduces a robust system for the detection and localisation of cars in images captured by air vehicles and satellites. The system adopts a sliding-window approach. It compromises a window-evaluation and a window-cl...

Detection of small targets, more specifically cars, in aerial images of urban scenes, has various applications in several domains, such as surveillance, military, remote sensing, and others. This is a tremendously challenging problem, mainly because of the significant interclass similarity among objects in urban environments, e.g., cars and certain...

In this paper, we propose a scare-aware pedestrian detection method assisted by a fast head-shoulder detection process to pre-estimate the candidate regions that may contain target pedestrians. Based on the observation that human head-shoulder regions share relatively robust features, we propose a head-shoulder detector using six aggregated feature...

An image fusion method that performs robustly for image sets heavily corrupted by noise is presented in this paper. The approach combines the advantages of two state-of-the-art fusion techniques, namely Independent Component Analysis (ICA) and Chebyshev Poly-nomial Analysis (CPA) fusion. Fusion using ICA performs well in transferring the salient fe...

Detection of cars in airborne images of typical urban areas has various applications in several domains, such as surveillance, military and remote sensing. It is a tremendously-challenging problem, mainly because of the significant inter-class similarity among various objects in urban environments. In this paper, a novel framework is introduced tha...

Analyzing and interpreting of thermograms have been increasingly employed in the diagnosis and monitoring of diseases thanks to its non-invasive, non-harmful nature and low cost. This paper presents a thermal image analysis system based on image registration for morphoea disease diagnosis. A novel system is proposed to improve the diagnosis and mon...

In this paper we propose a fast head-shoulder detector as a means to facilitating faster pedestrian detection. The proposed approach is based on the observation that human head-shoulder regions share relatively robust features. The purpose is to address the problem of high computational speed of the deformable part model (DPM) detector by selecting...

Building detection from two-dimensional high-resolution satellite images is a computer vision, photogrammetry, and remote sensing task that has arisen in the last decades with the advances in sensors technology and can be utilized in several applications that require the creation of urban maps or the study of urban changes. However, the variety of...

A novel adaptive image fusion method by using Chebyshev polynomial analysis (CPA), for applications in vegetation satellite imagery, is introduced in this paper. Fusion is a technique that enables the merging of two satellite cameras: panchromatic and multi-spectral, to produce higher quality satellite images to address agricurtural and vegetation...

Vision based human detection continuously attracts research interest since it is a topic of practical significance. The well-established Histogram of Oriented Gradients (HOG) human detector, though regarded as a reference for human detection, still suffers from the typical problem of the trade-off between precision and recall, relying on the thresh...

Pedestrian detection is an important image understanding problem with many potential applications. There has been little success in creating an algorithm which exhibits a high detection rate while keeping the false alarm in a relatively low rate. This paper presents a method designed to resolve this problem. The proposed method uses the Kinect or a...

Recognizing the imperative need for biodiversity protection, the convention on biological diversity (CBD) has recently established new targets towards 2020, the so-called Aichi targets, and updated proposed sets of indicators to quantitatively monitor the progress towards these targets. Remote sensing has been increasingly contributing to timely, a...

We investigate the problem of multi-view human gait recognition along any straight walking paths. It is observed that the gait appearance changes as the view changes while certain amount of correlated information exists among different views. Taking advantage of that type of correlation, a multi-view gait recognition method is proposed in this pape...

The guest editors would like to thank the authors for their excellent contributions and the reviewers for their help in improving the papers.

Vegetation height is a crucial factor in environmental studies, landscape analysis, and mapping applications. Its estimation may prove cost and resource demanding, e.g., employing light detection and ranging (LiDAR) data. This study presents a cost-effective framework for height estimation, built around texture analysis of a single very high-resolu...

In this work, the 3D flow estimation problem is formulated in the 4D spatiotemporal frequency domain, and it is shown that 3D motion manifests itself as energy concentration along hyper-planes in that domain. Based on this, the construction and use of appropriate directional multidimensional 'steerable' filters, which can extract directional energy...

Building segmentation from 2D images can be a very challenging task due to the variety of objects that appear in an urban environment. Many algorithms that attempt to automatically extract buildings from satellite images face serious problems and limitations. In this paper, we address some of these problems by applying a novel approach that is base...

An important cue that can assist towards an accurate building detection and segmentation is 3D information. Because of their height, buildings can easily be distinguished from the ground and small objects, allowing for their successful segmentation. Unfortunately, 3D knowledge is not always available, but there are ways to infer 3D information from...

Robust and efficient detection of cars in urban scenes has many useful applications. This paper introduces a framework for car detection from high-resolution satellite images, wherein a novel extended image descriptor is used to depict the geometric, spectral and colour distribution properties of cars. The proposed framework is based on a sliding-w...

The derivation of habitat maps is enhanced if land cover maps are used as basis for the mapping procedure. In this study, a supervised learning framework is proposed to perform object-based classification to General Habitat Categories. A Land Cover Classification System map is used as basis, and an approach to generate numerical features from the o...

This paper describes a novel approach of an adaptive fusion method by using Chebyshev polynomial analysis (CPA) for use in remote sensing vegetation imagery. Chebyshev polynomials have been effectively used in image fusion mainly in medium to high noise conditions, though its application is limited to heuristics. In this research, we have proposed...