This paper aims to develop a framework for generating synthetic data for the Swedish built environment to improve energy predictions and peak load management in heating systems. Heating and cooling are central aspects of reducing energy use in the residential sector. However, predicting the energy performance of buildings is currently a difficult task, partly due to energy use data from end-users being outdated and sometimes missing. In the end, we will contribute to future energy systems in a broader sense by developing a socio-technical and ethical methodological framework for working with synthetic energy data. We will map available data together with stakeholders' needs. Data will be collected and prepared as training data to develop and evaluate a model for synthetic data. The approach is interdisciplinary which will ensure the integration of socio-technical, ethical and gendered aspects of energy use and synthetic data.

Accurately evaluating the quality of bidirectional reflectance distribution function (BRDF) models is essential for photo-realistic rendering. Traditional BRDF-space metrics often employ numerical error measures that fail to capture perceptual differences evident in rendered images. In this paper, we introduce the first perceptually informed neural quality metric for BRDF evaluation that operates directly in BRDF space, eliminating the need for rendering during quality assessment. Our metric is implemented as a compact multi-layer perceptron (MLP), trained on a dataset of measured BRDFs supplemented with synthetically generated data and labelled using a perceptually validated image-space metric. The network takes as input paired samples of reference and approximated BRDFs and predicts their perceptual quality in terms of just-objectionable-difference (JOD) scores. We show that our neural metric achieves significantly higher correlation with human judgments than existing BRDF-space metrics. While its performance as a loss function for BRDF fitting remains limited, the proposed metric offers a perceptually grounded alternative for evaluating BRDF models.

The bidirectional reflectance distribution function (BRDF) describes the ratio of incoming radiance to outgoing radiance for all possible pairs of incoming and outgoing directions, defined over a spatial point. BRDF plays a key role in appearance modelling in computer graphics. Precise BRDF representation typically involves collecting millions of samples from incoming and outgoing directions, taking several hours or days of measurement using a gonioreflectometer. In this paper, we present an adaptive sampling framework for fast and accurate acquisition of BRDFs, where the number of measurements adapts to the complexity of the underlying BRDF function. We enhance the sampling efficiency of existing BRDF sampling techniques by accounting for the diverse reflectance properties of different materials. To achieve this, we categorise BRDFs in measured datasets into distinct clusters based on their sparsity and extract the necessary number of measurements for faithful reconstruction. Using a lightweight neural network, we predict the material's cluster from a single image, which allows us to determine the optimal sample count and sampling pattern, that is, the light/camera configuration. Our evaluation and analysis, compared to state-of-the-art methods, demonstrate a notable performance boost, particularly for challenging materials like specular BRDFs.

Publication of AI-generated synthetic structural data in data repositories is beginning to reveal the specific documentation elements that need to accompany synthetic datasets so as to ensure reproducibility and enable data reuse. This document identifies actions that research repositories can take to encourage users to provide AI-generated synthetic datasets with appropriate structure and documentation. The recommendations are specifically for AI generated data, not (for example) data produced using pre-configured models or missing data created by statistical inference. Additionally, this document discusses metadata/README elements for synthetic structured datasets (tabular and multi-modal) and not textual data from LLMs or images for computer vision. 

The document is the result of a workshop held on 23rd January 2025, with participants from the Swedish National Data Service, Linköping University and Manchester University. It also draws on survey responses about current practice from 17 data repositories and a review of existing metadata and README requirements. 

Material appearance is commonly modeled with the Bidirectional Reflectance Distribution Functions (BRDFs), which need to trade accuracy for complexity and storage cost. To investigate the current practices of BRDF modeling, we collect the first high dynamic range stereoscopic video dataset that captures the perceived quality degradation with respect to a number of parametric and non-parametric BRDF models. Our dataset shows that the current loss functions used to fit BRDF models, such as mean-squared error of logarithmic reflectance values, correlate poorly with the perceived quality of materials in rendered videos. We further show that quality metrics that compare rendered material samples give a significantly higher correlation with subjective quality judgments, and a simple Euclidean distance in the ITP color space (ΔEITP) shows the highest correlation. Additionally, we investigate the use of different BRDF-space metrics as loss functions for fitting BRDF models and find that logarithmic mapping is the most effective approach for BRDF-space loss functions.

Synthetic data is increasingly used as a substitute for real data due to ethical, legal, and logistical reasons. However, the rise of synthetic data also raises critical questions about its entanglement with the politics of classification and the reproduction of social norms and categories. This paper aims to problematize the use of synthetic data by examining how its production is intertwined with the maintenance of certain worldviews and classifications. We argue that synthetic data, like real data, is embedded with societal biases and power structures, leading to the reproduction of existing social inequalities. Through empirical examples, we demonstrate how synthetic data tends to highlight majority elements as the “normal” and minimize minority elements, and that the slight changes to the data structures that create synthetic data will also inevitably result in what we term “intersectional hallucinations.” These hallucinations are inherent to synthetic data and cannot be entirely eliminated without compromising the purpose of creating synthetic datasets. We contend that decisions about synthetic data involve determining which intersections are essential and which can be disregarded, a practice which will imbue these decisions with norms and values. Our study underscores the need for critical engagement with the mathematical and statistical choices in synthetic data production and advocates for careful consideration of the ontological and political implications of these choices during curatorial style production of synthetic structured data.

Hand in hand with the rapid development of machine learning, deep learning and generative AI algorithms and architectures, the graphics community has seen a remarkable evolution of novel techniques for material and appearance capture. Typically, these machine-learning-driven methods and technologies, in contrast to traditional techniques, rely on only a single or very few input images, while enabling the recovery of detailed, high-quality measurements of bi-directional reflectance distribution functions, as well as the corresponding spatially varying material properties, also known as Spatially Varying Bi-directional Reflectance Distribution Functions (SVBRDFs). Learning-based approaches for appearance capture will play a key role in the development of new technologies that will exhibit a significant impact on virtually all domains of graphics. Therefore, to facilitate future research, this State-of-the-Art Report (STAR) presents an in-depth overview of the state-of-the-art in machine-learning-driven material capture in general, and focuses on SVBRDF acquisition in particular, due to its importance in accurately modelling complex light interaction properties of real-world materials. The overview includes a categorization of current methods along with a summary of each technique, an evaluation of their functionalities, their complexity in terms of acquisition requirements, computational aspects and usability constraints. The STAR is concluded by looking forward and summarizing open challenges in research and development toward predictive and general appearance capture in this field. A complete list of the methods and papers reviewed in this survey is available at . Papers surveyed in this study with a focus on the extraction of BRDF or SVBRDF from a few measurements, classifying them according to their specific geometries and lighting conditions. Whole-scene refers to techniques that capture entire indoor or outdoor outside the scope of this survey. image

Traditional GAN (Generative Adversarial Network) architectures often reproduce biases present in their training data, leading to synthetic data that may unfairly impact certain subgroups. Past efforts to improve fairness in GANs usually target single demographic categories, like sex or race, but overlook intersectionality. Our approach addresses this gap by integrating an intersectionality framework with explainability techniques to identify and select problematic sensitive features. These insights are then used to develop intersectional fairness constraints integrated into the GAN training process. We aim to enhance fairness and maintain diverse subgroup representation by addressing intersections of multiple demographic attributes. Specifically, we adjusted the loss functions of two state-of-the-art GAN models for tabular data, including an intersectional demographic parity constraint. Our evaluations indicate that this approach significantly improves fairness in synthetically generated datasets. We compared the outcomes using Adult, and Diabetes datasets when considering the intersection of two sensitive features versus focusing on a single sensitive attribute, demonstrating the effectiveness of our method in capturing more complex biases.

As the problem of reconstructing high dynamic range (HDR) imagesfrom a single exposure has attracted much research effort, it isessential to provide a robust protocol and clear guidelines on howto evaluate and compare new methods. In this work, we comparedsix recent single image HDR reconstruction (SI-HDR) methodsin a subjective image quality experiment on an HDR display. Wefound that only two methods produced results that are, on average,more preferred than the unprocessed single exposure images. Whenthe same methods are evaluated using image quality metrics, astypically done in papers, the metric predictions correlate poorlywith subjective quality scores. The main reason is a significant toneand color difference between the reference and reconstructed HDRimages. To improve the predictions of image quality metrics, we propose correcting for the inaccuracies of the estimated cameraresponse curve before computing quality values. We further analyzethe sources of prediction noise when evaluating SI-HDR methodsand demonstrate that existing metrics can reliably predict onlylarge quality differences.