Zygmunt L. Szpak
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zygmunt L. Szpak has a total of 27 co-authors. Here are the top co-authors by number of shared publications:
- Wojciech Chojnacki
School of Computer and Mathematical Sciences, The University of Adelaide, Adelaide, Australia
14 shared publications Anton van Den Hengel
6 shared publications- Jules-Raymond Tapamo
Discipline of Electrical, Electronic and Computer Engineering, University of KwaZulu-Natal, Durban 4041, South Africa
3 shared publications - Andreas Cebulla
College of Human Sciences and Culture, Flinders University, Adelaide, Australia
2 shared publications - Aneta Neumann
Adelaide University
2 shared publications - Anton van Den Hengel
University of Adelaide, Adelaide, SA, Australia
2 shared publications
Latest Publications
- Open Access
- 15 July 2026
- Published by Wiley in Australian Journal of Rural Health
- Vol. 34 (4) , e70233
Acceptability and Utility of a Virtual Reality Diabetes‐Related Foot Disease Training Program for First Nations People: A Pragmatic Quasi‐Experimental Case Series Study
Objective: To test the utility and acceptability of a novel VR training tool for community health practitioners in assessing diabetes‐related foot disease in Aboriginal and Torres Strait Islander people. Methods: The VR program featured five virtual patient models and a range of interactive assessments covering essential clinical and triage skills. Design: Pragmatic quasi‐experimental case series. Setting: The VR training program was completed in collaboration with three Aboriginal Community‐Controlled Health Organisations in regional South Australia, with recruitment directed internally at each organisation. Participants: Forty‐one primary healthcare staff participated, of which 34 were matched on pre‐ and post‐VR surveys and 20 completed at least three training scenarios. Main Outcome Measures: Evaluation consisted of pre‐test self‐efficacy scores immediately prior to participants undergoing training, submission of post‐test self‐efficacy scores, the System Usability Scale, and Training Evaluation Inventory scores after completing all training modules. Results: Following completion of VR training, participants reported increased self‐efficacy scores ( n = 20) for performing key assessment and triage skills. User feedback ( n = 39) demonstrated that participants generally agreed that the training user experience was engaging, acceptable, and of value for clinical education. Conclusion: Community health practitioners generally found the VR training tool acceptable, useful, and increased their self‐confidence in assessing and managing diabetes‐related foot disease for Aboriginal and Torres Strait Islander people, highlighting the potential of VR to upskill community practitioners.
- 15 July 2026
- Published by Wiley in Australian Journal of Rural Health
- Vol. 34 (4) , e70233
Open AccessAcceptability and Utility of a Virtual Reality Diabetes‐Related Foot Disease Training Program for First Nations People: A Pragmatic Quasi‐Experimental Case Series Study
Objective: To test the utility and acceptability of a novel VR training tool for community health practitioners in assessing diabetes‐related foot disease in Aboriginal and Torres Strait Islander people. Methods: The VR program featured five virtual patient models and a range of interactive assessments covering essential clinical and triage skills. Design: Pragmatic quasi‐experimental case series. Setting: The VR training program was completed in collaboration with three Aboriginal Community‐Controlled Health Organisations in regional South Australia, with recruitment directed internally at each organisation. Participants: Forty‐one primary healthcare staff participated, of which 34 were matched on pre‐ and post‐VR surveys and 20 completed at least three training scenarios. Main Outcome Measures: Evaluation consisted of pre‐test self‐efficacy scores immediately prior to participants undergoing training, submission of post‐test self‐efficacy scores, the System Usability Scale, and Training Evaluation Inventory scores after completing all training modules. Results: Following completion of VR training, participants reported increased self‐efficacy scores ( n = 20) for performing key assessment and triage skills. User feedback ( n = 39) demonstrated that participants generally agreed that the training user experience was engaging, acceptable, and of value for clinical education. Conclusion: Community health practitioners generally found the VR training tool acceptable, useful, and increased their self‐confidence in assessing and managing diabetes‐related foot disease for Aboriginal and Torres Strait Islander people, highlighting the potential of VR to upskill community practitioners.
- 31 October 2023
- Published by Taylor & Francis
- p. 151-172
Workplace Relations with AI in Mind
Businesses increasingly use Artificial Intelligence (AI) to improve the delivery and the quality of products and services, but also to arrange workflows, recruit and performance-review employees. This chapter explores how these innovations have been – and are expected to be – changing workplace relations and workplace health and safety arrangements. It examines how systems of industrial relations, including Australia’s enterprise bargaining arrangements, are opening up to the challenges that AI may pose to traditional workplace relations.
- 31 October 2023
- Published by Taylor & Francis
- p. 151-172
Workplace Relations with AI in Mind
Businesses increasingly use Artificial Intelligence (AI) to improve the delivery and the quality of products and services, but also to arrange workflows, recruit and performance-review employees. This chapter explores how these innovations have been – and are expected to be – changing workplace relations and workplace health and safety arrangements. It examines how systems of industrial relations, including Australia’s enterprise bargaining arrangements, are opening up to the challenges that AI may pose to traditional workplace relations.
- Open Access
- 13 May 2022
- Published by Springer Nature in AI & SOCIETY
- Vol. 38 (2) , 919-935
Applying ethics to AI in the workplace: the design of a scorecard for Australian workplace health and safety
Artificial Intelligence (AI) is taking centre stage in economic growth and business operations alike. Public discourse about the practical and ethical implications of AI has mainly focussed on the societal level. There is an emerging knowledge base on AI risks to human rights around data security and privacy concerns. A separate strand of work has highlighted the stresses of working in the gig economy. This prevailing focus on human rights and gig impacts has been at the expense of a closer look at how AI may be reshaping traditional workplace relations and, more specifically, workplace health and safety. To address this gap, we outline a conceptual model for developing an AI Work Health and Safety (WHS) Scorecard as a tool to assess and manage the potential risks and hazards to workers resulting from AI use in a workplace. A qualitative, practice-led research study of AI adopters was used to generate and test a novel list of potential AI risks to worker health and safety. Risks were identified after cross-referencing Australian AI Ethics Principles and Principles of Good Work Design with AI ideation, design and implementation stages captured by the AI Canvas, a framework otherwise used for assessing the commercial potential of AI to a business. The unique contribution of this research is the development of a novel matrix itemising currently known or anticipated risks to the WHS and ethical aspects at each AI adoption stage.
- 13 May 2022
- Published by Springer Nature in AI & SOCIETY
- Vol. 38 (2) , 919-935
Open AccessApplying ethics to AI in the workplace: the design of a scorecard for Australian workplace health and safety
Artificial Intelligence (AI) is taking centre stage in economic growth and business operations alike. Public discourse about the practical and ethical implications of AI has mainly focussed on the societal level. There is an emerging knowledge base on AI risks to human rights around data security and privacy concerns. A separate strand of work has highlighted the stresses of working in the gig economy. This prevailing focus on human rights and gig impacts has been at the expense of a closer look at how AI may be reshaping traditional workplace relations and, more specifically, workplace health and safety. To address this gap, we outline a conceptual model for developing an AI Work Health and Safety (WHS) Scorecard as a tool to assess and manage the potential risks and hazards to workers resulting from AI use in a workplace. A qualitative, practice-led research study of AI adopters was used to generate and test a novel list of potential AI risks to worker health and safety. Risks were identified after cross-referencing Australian AI Ethics Principles and Principles of Good Work Design with AI ideation, design and implementation stages captured by the AI Canvas, a framework otherwise used for assessing the commercial potential of AI to a business. The unique contribution of this research is the development of a novel matrix itemising currently known or anticipated risks to the WHS and ethical aspects at each AI adoption stage.
- Open Access
Full explicit consistency constraints in uncalibrated multiple homography estimation
We reveal a complete set of constraints that need to be imposed on a set of 3-by-3 matrices to ensure that the matrices represent genuine homographies associated with multiple planes between two views. We also show how to exploit the constraints to obtain more accurate estimates of homography matrices between two views. Our study resolves a long-standing research question and provides a fresh perspective and a more in-depth understanding of the multiple homography estimation task.
Open AccessFull explicit consistency constraints in uncalibrated multiple homography estimation
We reveal a complete set of constraints that need to be imposed on a set of 3-by-3 matrices to ensure that the matrices represent genuine homographies associated with multiple planes between two views. We also show how to exploit the constraints to obtain more accurate estimates of homography matrices between two views. Our study resolves a long-standing research question and provides a fresh perspective and a more in-depth understanding of the multiple homography estimation task.
- 28 May 2019
- Published by Springer Nature
- p. 659-675
Full Explicit Consistency Constraints in Uncalibrated Multiple Homography Estimation
- 28 May 2019
- Published by Springer Nature
- p. 659-675
Full Explicit Consistency Constraints in Uncalibrated Multiple Homography Estimation
- Open Access
Determining ellipses from low-resolution images with a comprehensive image formation model
When determining the parameters of a parametric planar shape based on a single low-resolution image, common estimation paradigms lead to inaccurate parameter estimates. The reason behind poor estimation results is that standard estimation frameworks fail to model the image formation process at a sufficiently detailed level of analysis. We propose a new method for estimating the parameters of a planar elliptic shape based on a single photon-limited, low-resolution image. Our technique incorporates the effects of several elements - point-spread function, discretisation step, quantisation step, and photon noise - into a single cohesive and manageable statistical model. While we concentrate on the particular task of estimating the parameters of elliptic shapes, our ideas and methods have a much broader scope and can be used to address the problem of estimating the parameters of an arbitrary parametrically representable planar shape. Comprehensive experimental results on simulated and real imagery demonstrate that our approach yields parameter estimates with unprecedented accuracy. Furthermore, our method supplies a parameter covariance matrix as a measure of uncertainty for the estimated parameters, as well as a planar confidence region as a means for visualising the parameter uncertainty. The mathematical model developed in this paper may prove useful in a variety of disciplines which operate with imagery at the limits of resolution.
Open AccessDetermining ellipses from low-resolution images with a comprehensive image formation model
When determining the parameters of a parametric planar shape based on a single low-resolution image, common estimation paradigms lead to inaccurate parameter estimates. The reason behind poor estimation results is that standard estimation frameworks fail to model the image formation process at a sufficiently detailed level of analysis. We propose a new method for estimating the parameters of a planar elliptic shape based on a single photon-limited, low-resolution image. Our technique incorporates the effects of several elements - point-spread function, discretisation step, quantisation step, and photon noise - into a single cohesive and manageable statistical model. While we concentrate on the particular task of estimating the parameters of elliptic shapes, our ideas and methods have a much broader scope and can be used to address the problem of estimating the parameters of an arbitrary parametrically representable planar shape. Comprehensive experimental results on simulated and real imagery demonstrate that our approach yields parameter estimates with unprecedented accuracy. Furthermore, our method supplies a parameter covariance matrix as a measure of uncertainty for the estimated parameters, as well as a planar confidence region as a means for visualising the parameter uncertainty. The mathematical model developed in this paper may prove useful in a variety of disciplines which operate with imagery at the limits of resolution.
- 17 January 2019
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 36 (2) , 212-233
Determining ellipses from low-resolution images with a comprehensive image formation model
When determining the parameters of a parametric planar shape based on a single low-resolution image, common estimation paradigms lead to inaccurate parameter estimates. The reason behind poor estimation results is that standard estimation frameworks fail to model the image formation process at a sufficiently detailed level of analysis. We propose a new method for estimating the parameters of a planar elliptic shape based on a single photon-limited, low-resolution image. Our technique incorporates the effects of several elements—point-spread function, discretization step, quantization step, and photon noise—into a single cohesive and manageable statistical model. While we concentrate on the particular task of estimating the parameters of elliptic shapes, our ideas and methods have a much broader scope and can be used to address the problem of estimating the parameters of an arbitrary parametrically representable planar shape. Comprehensive experimental results on simulated and real imagery demonstrate that our approach yields parameter estimates with unprecedented accuracy. Furthermore, our method supplies a parameter covariance matrix as a measure of uncertainty for the estimated parameters, as well as a planar confidence region as a means for visualizing parameter uncertainty. The mathematical model developed in this paper may prove useful in a variety of disciplines that operate with imagery at the limits of resolution.
- 17 January 2019
- Published by Optica Publishing Group in Journal of the Optical Society of America A
- Vol. 36 (2) , 212-233
Determining ellipses from low-resolution images with a comprehensive image formation model
When determining the parameters of a parametric planar shape based on a single low-resolution image, common estimation paradigms lead to inaccurate parameter estimates. The reason behind poor estimation results is that standard estimation frameworks fail to model the image formation process at a sufficiently detailed level of analysis. We propose a new method for estimating the parameters of a planar elliptic shape based on a single photon-limited, low-resolution image. Our technique incorporates the effects of several elements—point-spread function, discretization step, quantization step, and photon noise—into a single cohesive and manageable statistical model. While we concentrate on the particular task of estimating the parameters of elliptic shapes, our ideas and methods have a much broader scope and can be used to address the problem of estimating the parameters of an arbitrary parametrically representable planar shape. Comprehensive experimental results on simulated and real imagery demonstrate that our approach yields parameter estimates with unprecedented accuracy. Furthermore, our method supplies a parameter covariance matrix as a measure of uncertainty for the estimated parameters, as well as a planar confidence region as a means for visualizing parameter uncertainty. The mathematical model developed in this paper may prove useful in a variety of disciplines that operate with imagery at the limits of resolution.
- Open Access
- 1 July 2017
- Published by Association for Computing Machinery (ACM)
- p. 817-824
Evolutionary image composition using feature covariance matrices
Evolutionary algorithms have recently been used to create a wide range of artistic work. In this paper, we propose a new approach for the composition of new images from existing ones, that retain some salient features of the original images. We introduce evolutionary algorithms that create new images based on a fitness function that incorporates feature covariance matrices associated with different parts of the images. This approach is very flexible in that it can work with a wide range of features and enables targeting specific regions in the images. For the creation of the new images, we propose a population-based evolutionary algorithm with mutation and crossover operators based on random walks. Our experimental results reveal a spectrum of aesthetically pleasing images that can be obtained with the aid of our evolutionary process.
- 1 July 2017
- Published by Association for Computing Machinery (ACM)
- p. 817-824
Open AccessEvolutionary image composition using feature covariance matrices
Evolutionary algorithms have recently been used to create a wide range of artistic work. In this paper, we propose a new approach for the composition of new images from existing ones, that retain some salient features of the original images. We introduce evolutionary algorithms that create new images based on a fitness function that incorporates feature covariance matrices associated with different parts of the images. This approach is very flexible in that it can work with a wide range of features and enables targeting specific regions in the images. For the creation of the new images, we propose a population-based evolutionary algorithm with mutation and crossover operators based on random walks. Our experimental results reveal a spectrum of aesthetically pleasing images that can be obtained with the aid of our evolutionary process.
- Open Access
Evolutionary Image Composition Using Feature Covariance Matrices
Evolutionary algorithms have recently been used to create a wide range of artistic work. In this paper, we propose a new approach for the composition of new images from existing ones, that retain some salient features of the original images. We introduce evolutionary algorithms that create new images based on a fitness function that incorporates feature covariance matrices associated with different parts of the images. This approach is very flexible in that it can work with a wide range of features and enables targeting specific regions in the images. For the creation of the new images, we propose a population-based evolutionary algorithm with mutation and crossover operators based on random walks. Our experimental results reveal a spectrum of aesthetically pleasing images that can be obtained with the aid of our evolutionary process.
Open AccessEvolutionary Image Composition Using Feature Covariance Matrices
Evolutionary algorithms have recently been used to create a wide range of artistic work. In this paper, we propose a new approach for the composition of new images from existing ones, that retain some salient features of the original images. We introduce evolutionary algorithms that create new images based on a fitness function that incorporates feature covariance matrices associated with different parts of the images. This approach is very flexible in that it can work with a wide range of features and enables targeting specific regions in the images. For the creation of the new images, we propose a population-based evolutionary algorithm with mutation and crossover operators based on random walks. Our experimental results reveal a spectrum of aesthetically pleasing images that can be obtained with the aid of our evolutionary process.
- 1 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
A Study of the Region Covariance Descriptor: Impact of Feature Selection and Image Transformations
We analyse experimentally the region covariance descriptor which has proven useful in numerous computer vision applications. The properties of the descriptor--despite its widespread deployment--are not well understood or documented. In an attempt to uncover key attributes of the descriptor, we characterise the interdependence between the choice of features and distance measures through a series of meticulously designed and performed experiments. Our results paint a rather complex picture and underscore the necessity for more extensive empirical and theoretical work. In light of our findings, there is reason to believe that the region covariance descriptor will prove useful for methods that perform image super-resolution, deblurring, and denoising based on matching and retrieval of image patches from an image dictionary.
- 1 November 2015
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-8
A Study of the Region Covariance Descriptor: Impact of Feature Selection and Image Transformations
We analyse experimentally the region covariance descriptor which has proven useful in numerous computer vision applications. The properties of the descriptor--despite its widespread deployment--are not well understood or documented. In an attempt to uncover key attributes of the descriptor, we characterise the interdependence between the choice of features and distance measures through a series of meticulously designed and performed experiments. Our results paint a rather complex picture and underscore the necessity for more extensive empirical and theoretical work. In light of our findings, there is reason to believe that the region covariance descriptor will prove useful for methods that perform image super-resolution, deblurring, and denoising based on matching and retrieval of image patches from an image dictionary.