Gyorgy Denes

Dr. Gyorgy Denes

orcid

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
20 November 2017 in ACM Transactions on Graphics
Verified
Journal Article
Open access
PDF file
GEGabriel Eilertsen
Gabriel Eilertsen
JKJoel Kronander
Joel Kronander
GDGyorgy Denes
Gyorgy Denes
RMRafał K. Mantiuk
Rafał K. Mantiuk
JUJonas Unger
Jonas Unger
Camera sensors can only capture a limited range of luminance simultaneously, and in order to create high dynamic range (HDR) images a set of different exposures are typically combined. In this paper we address the problem of predicting information that have been lost in saturated image areas, in order to enable HDR reconstruction from a single exposure. We show that this problem is well-suited for deep learning algorithms, and propose a deep convolutional neural network (CNN) that is specifically designed taking into account the challenges in predicting HDR values. To train the CNN we gather a large dataset of HDR images, which we augment by simulating sensor saturation for a range of cameras. To further boost robustness, we pre-train the CNN on a simulated HDR dataset created from a subset of the MIT Places database. We demonstrate that our approach can reconstruct high-resolution visually convincing HDR results in a wide range of situations, and that it generalizes well to reconstruction of images captured with arbitrary and low-end cameras that use unknown camera response functions and post-processing. Furthermore, we compare to existing methods for HDR expansion, and show high quality results also for image based lighting. Finally, we evaluate the results in a subjective experiment performed on an HDR display. This shows that the reconstructed HDR images are visually convincing, with large improvements as compared to existing methods.

18 February 2019 in IEEE Transactions on Visualization and Computer Graphics
Verified
Journal Article
Closed access
GDGyorgy Denes
Gyorgy Denes
KMKuba Maruszczyk
Kuba Maruszczyk
GAGeorge Ash
George Ash
RMRafal K. Mantiuk
Rafal K. Mantiuk
Rendering in virtual reality (VR) requires substantial computational power to generate 90 frames per second at high resolution with good-quality antialiasing. The video data sent to a VR headset requires high bandwidth, achievable only on dedicated links. In this paper we explain how rendering requirements and transmission bandwidth can be reduced using a conceptually simple technique that integrates well with existing rendering pipelines. Every even-numbered frame is rendered at a lower resolution, and every odd-numbered frame is kept at high resolution but is modified in order to compensate for the previous loss of high spatial frequencies. When the frames are seen at a high frame rate, they are fused and perceived as high-resolution and high-frame-rate animation. The technique relies on the limited ability of the visual system to perceive high spatio-temporal frequencies. Despite its conceptual simplicity, correct execution of the technique requires a number of non-trivial steps: display photometric temporal response must be modeled, flicker and motion artifacts must be avoided, and the generated signal must not exceed the dynamic range of the display. Our experiments, performed on a high-frame-rate LCD monitor and OLED-based VR headsets, explore the parameter space of the proposed technique and demonstrate that its perceived quality is indistinguishable from full-resolution rendering. The technique is an attractive alternative to reprojection and resolution reduction of all frames.

28 January 2023 in Computers and Education: Artificial Intelligence
Verified
Journal Article
Open access
GDGyorgy Denes
Gyorgy Denes
The COVID-19 pandemic has created significant challenges for UK schools, but a time of cancelled exams and uncertainty around future examinations can provide opportunities to explore novel assessment methods. Hence, the 2020 proposal of the Ofqual algorithm which combines teachers' estimated grades and schools' historical performance seemed timely. However, the algorithmically calculated grades resulted in a public backlash and withdrawal of the proposal. While the failed Ofqual algorithm could be considered an example of AI, we do not yet have a thorough understanding of its numerical accuracy and how it performs in comparison to other AI models. This paper investigates this novel application: the potential use of a range of AI models as assessment tools in a selective, independent, secondary school in England. The following questions were examined: (1) how accurate are modern AI models in predicting GCSE exam grades? (2) what are the differences in model accuracy across subjects and can these be explained by qualitative differences in teachers' grading practices? Results indicate that while models yield acceptable mean absolute errors, individual mispredictions can be larger than desired. Subject differences highlighted that grading subjectivity is less significant in science, technology, engineering, and maths (STEM) subjects, which could explain why objective models fail to predict non-STEM grades more frequently. In summary, numerical results indicate that grade prediction could be an interesting novel application of AI, but more research is needed to reduce outliers.

17 February 2024 in Computers
Verified
Journal Article
Open access
PDF file
AJAmaan Ibraheem Jamil
Amaan Ibraheem Jamil
GDGyorgy Denes
Gyorgy Denes
Over 300 million people who live with color vision deficiency (CVD) have a decreased ability to distinguish between colors, limiting their ability to interact with websites and software packages. User-interface designers have taken various approaches to tackle the issue, with most offering a high-contrast mode. The Web Content Accessibility Guidelines (WCAG) outline some best practices for maintaining accessibility that have been adopted and recommended by several governments; however, it is currently uncertain how this impacts perceived user functionality and if this could result in a reduced aesthetic look. In the absence of subjective data, we aim to investigate how a CVD observer might rate the functionality and aesthetics of existing UIs. However, the design of a comparative study of CVD vs. non-CVD populations is inherently hard; therefore, we build on the successful field of physiologically based CVD models and propose a novel simulation-based experimental protocol, where non-CVD observers rate the relative aesthetics and functionality of screenshots of 20 popular websites as seen in full color vs. with simulated CVD. Our results show that relative aesthetics and functionality correlate positively and that an operating-system-wide high-contrast mode can reduce both aesthetics and functionality. While our results are only valid in the context of simulated CVD screenshots, the approach has the benefit of being easily deployable, and can help to spot a number of common pitfalls in production. Finally, we propose a AAA–A classification of the interfaces we analyzed.

30 January 2024 in ACM Transactions on Applied Perception
Verified
Journal Article
Open access
PEPontus Ebelin
Pontus Ebelin
GDGyorgy Denes
Gyorgy Denes
TATomas Akenine-Möller
Tomas Akenine-Möller
KÅKalle Åström
Kalle Åström
MOMagnus Oskarsson
Magnus Oskarsson
WMWilliam H. McIlhagga
William H. McIlhagga
Edge detection is an important process in human visual processing. However, as far as we know, few attempts have been made to map the temporal edge detection filters in human vision. To that end, we devised a user study and collected data from which we derived estimates of human temporal edge detection filters based on three different models, including the derivative of the infinite symmetric exponential function and temporal contrast sensitivity function. We analyze our findings using several different methods, including extending the filter to higher frequencies than were shown during the experiment. In addition, we show a proof of concept that our filter may be used in spatiotemporal image quality metrics by incorporating it into a flicker detection pipeline.

24 February 2024 in arXiv
Verified
Preprint
Open access
PDF file
AJAmaan Ibraheem Jamil
Amaan Ibraheem Jamil
GDGyorgy Denes
Gyorgy Denes
Over 300 million people who live with color vision deficiency (CVD) have a decreased ability to distinguish between colors, limiting their ability to interact with websites and software packages. User interface designers have taken various approaches to tackle the issue with most offering a high contrast mode. The Web Content Accessibility Guidelines (WCAG) outline some best practices for maintaining accessibility that have been adopted and recommended by several governments; however, it is currently uncertain how this impacts perceived user functionality and if this could result in a reduced aesthetic look. In the absence of subjective data, we aim to investigate how a CVD observer might rate the functionality and aesthetics of existing UIs. However, the design of a comparative study of CVD vs. non-CVD populations is inherently hard, therefore we build on the successful field of physiologically-based CVD models, and propose a novel simulation-based experimental protocol, where non-CVD observers rate the relative aesthetics and functionality of screenshots of 20 popular websites as seen in full color vs. with simulated CVD. Our results show that relative aesthetics and functionality correlate positively and that an operating-system-wide high contrast mode can reduce both aesthetics and functionality. While our results are only valid in the context of simulated CVD screenshots, the approach has the benefit of being easily deployable, and can help to spot a number of common pitfalls in production. Finally, we propose a AAA-A classification of the interfaces we analyzed.