arXiv subjects

Joan Lasenby

Publications and source records attributed to Joan Lasenby.

At least 37 records · Page 2Linked to original sources

arXiv2022

SPIN: Simplifying Polar Invariance for Neural networks Application to vision-based irradiance forecasting

Translational invariance induced by pooling operations is an inherent property of convolutional neural networks, which facilitates numerous computer vision tasks such as classification. Yet to leverage rotational invariant tasks, convolutional architectures require specific rotational invariant layers or extensive data augmentation to learn from diverse rotated versions of a given spatial configuration. Unwrapping the image into its polar coordinates provides a more explicit representation to train a convolutional architecture as the rotational invariance becomes translational, hence the visually distinct but otherwise equivalent rotated versions of a given scene can be learnt from a single image. We show with two common vision-based solar irradiance forecasting challenges (i.e. using ground-taken sky images or satellite images), that this preprocessing step significantly improves prediction results by standardising the scene representation, while decreasing training time by a factor of 4 compared to augmenting data with rotations. In addition, this transformation magnifies the area surrounding the centre of the rotation, leading to more accurate short-term irradiance predictions.

cs.CV↗

arXiv2022

Closed-form solutions for the inverse kinematics of serial robots using conformal geometric algebra

This work addresses the inverse kinematics of serial robots using conformal geometric algebra. Classical approaches include either the use of homogeneous matrices, which entails high computational cost and execution time or the development of particular geometric strategies that cannot be generalized to arbitrary serial robots. In this work, we present a compact, elegant and intuitive formulation of robot kinematics based on conformal geometric algebra that provides a suitable framework for the closed-form resolution of the inverse kinematic problem for manipulators with a spherical wrist. For serial robots of this kind, the inverse kinematics problem can be split in two subproblems: the position and orientation problems. The latter is solved by appropriately splitting the rotor that defines the target orientation into three simpler rotors, while the former is solved by developing a geometric strategy for each combination of prismatic and revolute joints that forms the position part of the robot. Finally, the inverse kinematics of 7 DoF redundant manipulators with a spherical wrist is solved by extending the geometric solutions obtained in the non-redundant case.

cs.RO↗

arXiv2021

Advancing COVID-19 Diagnosis with Privacy-Preserving Collaboration in Artificial Intelligence

Artificial intelligence (AI) provides a promising substitution for streamlining COVID-19 diagnoses. However, concerns surrounding security and trustworthiness impede the collection of large-scale representative medical data, posing a considerable challenge for training a well-generalised model in clinical practices. To address this, we launch the Unified CT-COVID AI Diagnostic Initiative (UCADI), where the AI model can be distributedly trained and independently executed at each host institution under a federated learning framework (FL) without data sharing. Here we show that our FL model outperformed all the local models by a large yield (test sensitivity /specificity in China: 0.973/0.951, in the UK: 0.730/0.942), achieving comparable performance with a panel of professional radiologists. We further evaluated the model on the hold-out (collected from another two hospitals leaving out the FL) and heterogeneous (acquired with contrast materials) data, provided visual explanations for decisions made by the model, and analysed the trade-offs between the model performance and the communication costs in the federated training process. Our study is based on 9,573 chest computed tomography scans (CTs) from 3,336 patients collected from 23 hospitals located in China and the UK. Collectively, our work advanced the prospects of utilising federated learning for privacy-preserving AI in digital health.

cs.AI↗

arXiv2021

Unsupervised Point Cloud Pre-Training via Occlusion Completion

We describe a simple pre-training approach for point clouds. It works in three steps: 1. Mask all points occluded in a camera view; 2. Learn an encoder-decoder model to reconstruct the occluded points; 3. Use the encoder weights as initialisation for downstream point cloud tasks. We find that even when we construct a single pre-training dataset (from ModelNet40), this pre-training method improves accuracy across different datasets and encoders, on a wide range of downstream tasks. Specifically, we show that our method outperforms previous pre-training methods in object classification, and both part-based and semantic segmentation tasks. We study the pre-trained features and find that they lead to wide downstream minima, have high transformation invariance, and have activations that are highly correlated with part labels. Code and data are available at: https://github.com/hansen7/OcCo

cs.CV↗

arXiv2021

Singularities of serial robots: Identification and distance computation using geometric algebra

The singularities of serial robotic manipulators are those configurations in which the robot loses the ability to move in at least one direction. Hence, their identification is fundamental to enhance the performance of current control and motion planning strategies. While classical approaches entail the computation of the determinant of either a 6x n or nxn matrix for an n degrees of freedom serial robot, this work addresses a novel singularity identification method based on modelling the twists defined by the joint axes of the robot as vectors of the six-dimensional and three-dimensional geometric algebras. In particular, it consists of identifying which configurations cause the exterior product of these twists to vanish. In addition, since rotors represent rotations in geometric algebra, once these singularities have been identified, a distance function is defined in the configuration space C such that its restriction to the set of singular configurations S allows us to compute the distance of any configuration to a given singularity. This distance function is used to enhance how the singularities are handled in three different scenarios, namely motion planning, motion control and bilateral teleoperation.

cs.RO↗

arXiv2021

Redesigning Fully Convolutional DenseUNets for Large Histopathology Images

The automated segmentation of cancer tissue in histopathology images can help clinicians to detect, diagnose, and analyze such disease. Different from other natural images used in many convolutional networks for benchmark, histopathology images can be extremely large, and the cancerous patterns can reach beyond 1000 pixels. Therefore, the well-known networks in the literature were never conceived to handle these peculiarities. In this work, we propose a Fully Convolutional DenseUNet that is particularly designed to solve histopathology problems. We evaluated our network in two public pathology datasets published as challenges in the recent MICCAI 2019: binary segmentation in colon cancer images (DigestPath2019), and multi-class segmentation in prostate cancer images (Gleason2019), achieving similar and better results than the winners of the challenges, respectively. Furthermore, we discussed some good practices in the training setup to yield the best performance and the main challenges in these histopathology datasets.

cs.CV↗

arXiv2021

Rotaflip: A New CNN Layer for Regularization and Rotational Invariance in Medical Images

Regularization in convolutional neural networks (CNNs) is usually addressed with dropout layers. However, dropout is sometimes detrimental in the convolutional part of a CNN as it simply sets to zero a percentage of pixels in the feature maps, adding unrepresentative examples during training. Here, we propose a CNN layer that performs regularization by applying random rotations of reflections to a small percentage of feature maps after every convolutional layer. We prove how this concept is beneficial for images with orientational symmetries, such as in medical images, as it provides a certain degree of rotational invariance. We tested this method in two datasets, a patch-based set of histopathology images (PatchCamelyon) to perform classification using a generic DenseNet, and a set of specular microscopy images of the corneal endothelium to perform segmentation using a tailored U-net, improving the performance in both cases.

eess.IV↗

arXiv2021

Benchmarking of Deep Learning Irradiance Forecasting Models from Sky Images -- an in-depth Analysis

A number of industrial applications, such as smart grids, power plant operation, hybrid system management or energy trading, could benefit from improved short-term solar forecasting, addressing the intermittent energy production from solar panels. However, current approaches to modelling the cloud cover dynamics from sky images still lack precision regarding the spatial configuration of clouds, their temporal dynamics and physical interactions with solar radiation. Benefiting from a growing number of large datasets, data driven methods are being developed to address these limitations with promising results. In this study, we compare four commonly used Deep Learning architectures trained to forecast solar irradiance from sequences of hemispherical sky images and exogenous variables. To assess the relative performance of each model, we used the Forecast Skill metric based on the smart persistence model, as well as ramp and time distortion metrics. The results show that encoding spatiotemporal aspects of the sequence of sky images greatly improved the predictions with 10 min ahead Forecast Skill reaching 20.4% on the test year. However, based on the experimental data, we conclude that, with a common setup, Deep Learning models tend to behave just as a 'very smart persistence model', temporally aligned with the persistence model while mitigating its most penalising errors. Thus, despite being captured by the sky cameras, models often miss fundamental events causing large irradiance changes such as clouds obscuring the sun. We hope that our work will contribute to a shift of this approach to irradiance forecasting, from reactive to anticipatory.

cs.CV↗

arXiv2020

Neural Random Subspace

The random subspace method, known as the pillar of random forests, is good at making precise and robust predictions. However, there is not a straightforward way yet to combine it with deep learning. In this paper, we therefore propose Neural Random Subspace (NRS), a novel deep learning based random subspace method. In contrast to previous forest methods, NRS enjoys the benefits of end-to-end, data-driven representation learning, as well as pervasive support from deep learning software and hardware platforms, hence achieving faster inference speed and higher accuracy. Furthermore, as a non-linear component to be encoded into Convolutional Neural Networks (CNNs), NRS learns non-linear feature representations in CNNs more efficiently than previous higher-order pooling methods, producing good results with negligible increase in parameters, floating point operations (FLOPs) and real running time. Compared with random subspaces, random forests and gradient boosting decision trees (GBDTs), NRS achieves superior performance on 35 machine learning datasets. Moreover, on both 2D image and 3D point cloud recognition tasks, integration of NRS with CNN architectures achieves consistent improvements with minor extra cost. Code is available at https://github.com/CupidJay/NRS_pytorch.

cs.LG↗

arXiv2020

Convolutional Neural Networks applied to sky images for short-term solar irradiance forecasting

Despite the advances in the field of solar energy, improvements of solar forecasting techniques, addressing the intermittent electricity production, remain essential for securing its future integration into a wider energy supply. A promising approach to anticipate irradiance changes consists of modeling the cloud cover dynamics from ground taken or satellite images. This work presents preliminary results on the application of deep Convolutional Neural Networks for 2 to 20 min irradiance forecasting using hemispherical sky images and exogenous variables. We evaluate the models on a set of irradiance measurements and corresponding sky images collected in Palaiseau (France) over 8 months with a temporal resolution of 2 min. To outline the learning of neural networks in the context of short-term irradiance forecasting, we implemented visualisation techniques revealing the types of patterns recognised by trained algorithms in sky images. In addition, we show that training models with past samples of the same day improves their forecast skill, relative to the smart persistence model based on the Mean Square Error, by around 10% on a 10 min ahead prediction. These results emphasise the benefit of integrating previous same-day data in short-term forecasting. This, in turn, can be achieved through model fine tuning or using recurrent units to facilitate the extraction of relevant temporal features from past data.

cs.CV↗

arXiv2018

The unreasonable effectiveness of the forget gate

Given the success of the gated recurrent unit, a natural question is whether all the gates of the long short-term memory (LSTM) network are necessary. Previous research has shown that the forget gate is one of the most important gates in the LSTM. Here we show that a forget-gate-only version of the LSTM with chrono-initialized biases, not only provides computational savings but outperforms the standard LSTM on multiple benchmark datasets and competes with some of the best contemporary models. Our proposed network, the JANET, achieves accuracies of 99% and 92.5% on the MNIST and pMNIST datasets, outperforming the standard LSTM which yields accuracies of 98.5% and 91%.

cs.NE↗

arXiv2018

Techniques for visualizing LSTMs applied to electrocardiograms

This paper explores four different visualization techniques for long short-term memory (LSTM) networks applied to continuous-valued time series. On the datasets analysed, we find that the best visualization technique is to learn an input deletion mask that optimally reduces the true class score. With a specific focus on single-lead electrocardiograms from the MIT-BIH arrhythmia dataset, we show that salient input features for the LSTM classifier align well with medical theory.

stat.ML↗

arXiv2017

Using geometric algebra to represent curvature in shell theory with applications to Starling resistors

We present a novel application of rotors in geometric algebra to represent the change of curvature tensor, that is used in shell theory as part of the constitutive law. We introduce a new decomposition of the change of curvature tensor, which has explicit terms for changes of curvature due to initial curvature combined with strain, and changes in rotation over the surface. We use this decomposition to perform a scaling analysis of the relative importance of bending and stretching in flexible tubes undergoing self excited oscillations. These oscillations have relevance to the lung, in which it is believed that they are responsible for wheezing. The new analysis is necessitated by the fact that the working fluid is air, compared to water in most previous work. We use stereographic imaging to empirically measure the relative importance of bending and stretching energy in observed self excited oscillations. This enables us to validate our scaling analysis. We show that bending energy is dominated by stretching energy, and the scaling analysis makes clear that this will remain true for tubes in the airways of the lung.

math-ph↗

arXiv2017

Bayesian LSTMs in medicine

The medical field stands to see significant benefits from the recent advances in deep learning. Knowing the uncertainty in the decision made by any machine learning algorithm is of utmost importance for medical practitioners. This study demonstrates the utility of using Bayesian LSTMs for classification of medical time series. Four medical time series datasets are used to show the accuracy improvement Bayesian LSTMs provide over standard LSTMs. Moreover, we show cherry-picked examples of confident and uncertain classifications of the medical time series. With simple modifications of the common practice for deep learning, significant improvements can be made for the medical practitioner and patient.

stat.ML↗

arXiv2017

The Elastic Theory of Shells using Geometric Algebra

We present a novel derivation of the elastic theory of shells. We use the language of Geometric algebra, which allows us to express the fundamental laws in component-free form, thus aiding physical interpretation. It also provides the tools to express equations in an arbitrary coordinate system, which enhances their usefulness. The role of moments and angular velocity, and the apparent use by previous authors of an unphysical angular velocity, has been clarified through the use of a bivector representation. In the linearised theory, clarification of previous coordinate conventions which have been the cause of confusion, is provided, and the introduction of prior strain into the linearised theory of shells is made possible.

math-ph↗

arXiv2017

Geometric algebra and an acoustic space time for propagation in non-uniform flow

This study aims to make use of two concepts in the field of aeroacoustics; an analogy with relativity, and Geometric Algebra. The analogy with relativity has been investigated in physics and cosmology, but less has been done to use this work in the field of aeroacoustics. Despite being successfully applied to a variety of fields, Geometric Algebra has yet to be applied to acoustics. Our aim is to apply these concepts first to a simple problem in aeroacoustics, sound propagation in uniform flow, and the more general problem of acoustic propagation in non-uniform flows. By using Geometric Algebra we are able to provide a simple geometric interpretation to a transformation commonly used to solve for sound fields in uniform flow. We are then able to extend this concept to an acoustic space-time applicable to irrotational, barotropic background flows. This geometrical framework is used to naturally derive the requirements that must be satisfied by the background flow in order for us to be able to solve for sound propagation in the non-uniform flow using the simple wave equation. We show that this is not possible in the most general situation, and provide an explicit expression that must be satisfied for the transformation to exist. We show that this requirement is automatically satisfied if the background flow is incompressible or uniform, and for both these cases derive an explicit transformation. In addition to a new physical interpretation for the transformation, we show that unlike previous investigations, our work is applicable to any frequency.

physics.flu-dyn↗

arXiv2015

An acoustic space-time and the Lorentz transformation in aeroacoustics

In this paper we introduce concepts from relativity and geometric algebra to aeroacoustics. We do this using an acoustic space-time transformation within the framework of sound propagation in uniform flows. By using Geometric Algebra we are able to provide a simple geometric interpretation to the space-time transformation, and are able to give neat and lucid derivations of the free-field Green's function for the convected wave equation and the Doppler shift for a stationary observer and a source in uniform rectilinear motion in a uniform flow.

physics.flu-dyn↗

arXiv2014

Single camera pose estimation using Bayesian filtering and Kinect motion priors

Traditional approaches to upper body pose estimation using monocular vision rely on complex body models and a large variety of geometric constraints. We argue that this is not ideal and somewhat inelegant as it results in large processing burdens, and instead attempt to incorporate these constraints through priors obtained directly from training data. A prior distribution covering the probability of a human pose occurring is used to incorporate likely human poses. This distribution is obtained offline, by fitting a Gaussian mixture model to a large dataset of recorded human body poses, tracked using a Kinect sensor. We combine this prior information with a random walk transition model to obtain an upper body model, suitable for use within a recursive Bayesian filtering framework. Our model can be viewed as a mixture of discrete Ornstein-Uhlenbeck processes, in that states behave as random walks, but drift towards a set of typically observed poses. This model is combined with measurements of the human head and hand positions, using recursive Bayesian estimation to incorporate temporal information. Measurements are obtained using face detection and a simple skin colour hand detector, trained using the detected face. The suggested model is designed with analytical tractability in mind and we show that the pose tracking can be Rao-Blackwellised using the mixture Kalman filter, allowing for computational efficiency while still incorporating bio-mechanical properties of the upper body. In addition, the use of the proposed upper body model allows reliable three-dimensional pose estimates to be obtained indirectly for a number of joints that are often difficult to detect using traditional object recognition strategies. Comparisons with Kinect sensor results and the state of the art in 2D pose estimation highlight the efficacy of the proposed approach.

cs.CV↗