An individualised (HRTF) is very important for creating realistic (VR) and (AR) environments. However, acoustically measuring high-quality HRTFs requires expensive equipment an acoustic lab setting. To overcome these limitations to make this measurement more efficient HRTF upsampling has been exploited in the past where a high-resolution created from low-resolution one. This paper demonstrates how (GAN) can be applied upsampling. We propose novel approach that transforms data direct use with...

Materials characterization is fundamental to our understanding of lithium ion battery electrodes and their performance limitations. Advances in laboratory-based techniques have yielded powerful insights into the structure–function relationship electrodes, yet there still far go. Further improvements rely, part, on gaining a deeper complex physical heterogeneities materials. However, practical limitations inhibit ability combine data directly. For example, some are destructive, thus...

Abstract Modelling the impact of a material's mesostructure on device level performance typically requires access to 3D image data containing all relevant information define geometry simulation domain. This must include sufficient contrast between phases, be high enough resolution capture key details, but also have large field‐of‐view representative material in general. It is rarely possible obtain with these properties from single imaging technique. In this paper, we present method for...

Abstract 3D microstructural datasets are commonly used to define the geometrical domains in finite element modelling. This has proven a useful tool for understanding how complex material systems behave under applied stresses, temperatures and chemical conditions. However, imaging of materials is challenging number reasons, including limited field view, low resolution difficult sample preparation. Recently, machine learning method, SliceGAN , was developed statistically generate arbitrary...

Segmentation is the assigning of a semantic class to every pixel in an image and prerequisite for various statistical analysis tasks materials science, like phase quantification, physics simulations or morphological characterisation.The wide range length scales, imaging techniques studied science means any segmentation algorithm must generalise unseen data support abstract, user-defined classes.Trainable popular interactive paradigm where classifier trained map from features user drawn...

Applying ultrathin MgO as the top component of a ZnO/MgO electron transport layer enhances performance organic photovoltaics.

TauFactor 2 is an open-source, GPU accelerated microstructural analysis tool for extracting metrics from voxel based data, including transport properties such as the touristy factor.Tortuosity factor, , a material parameter that defines reduction in arising arrangement of phases multiphase medium (see Figure 1).As shown Equation 1, effective coefficient material, eff can be calculated intrinsic coefficient, volume fraction, and (Cooper et al., 2016) (note, this value should not squared...

Headphones-based spatial audio simulations rely on Head-related Transfer Functions (HRTFs) in order to reconstruct the sound field at entrance of listener’s ears. A HRTF is strongly dependent specific anatomical structures, and it has been shown that virtual sounds recreated with someone else’s result worse localisation accuracy, as well altering other subjective measures such externalisation realism. Acoustic measurements filtering effects generated by ears, head torso proven be one most...

Modelling lithium-ion battery behavior is essential for performance prediction and design improvement. However, this task challenging due to processes spanning many length scales, leading computationally expensive models. Reduced order models have been developed address this, assuming a “separation of scales” between micro- macroscales. This study compares two approaches: direct microstructure-resolved 3D domain electrochemical modelling simplified 1D homogenized model, similar the...

We present a novel inpainting algorithm for microstructural image data using generative adversarial networks. This enables fast artefact removal via simple graphical user interface.

Segmentation is the assigning of a semantic class to every pixel in an image and prerequisite for various statistical analysis tasks materials science, like phase quantification, physics simulations or morphological characterization. The wide range length scales, imaging techniques studied science means any segmentation algorithm must generalise unseen data support abstract, user-defined classes. Trainable popular interactive paradigm where classifier trained map from features user drawn...

Imaging It is hard to obtain mesostructural image data with sufficient contrast, resolution, and field-of-view from a single imaging technique. In article number 2202407, Amir Dahari, Samuel J. Cooper co-workers present method for combining information pairs of complementary 2D 3D techniques in order accurately reconstruct the desired volumes. The uses generative adversarial networks perform super-resolution, style-transfer, dimensionality expansion.

An individualised head-related transfer function (HRTF) is very important for creating realistic virtual reality (VR) and augmented (AR) environments. However, acoustically measuring high-quality HRTFs requires expensive equipment an acoustic lab setting. To overcome these limitations to make this measurement more efficient HRTF upsampling has been exploited in the past where a high-resolution created from low-resolution one. This paper demonstrates how generative adversarial networks (GANs)...

3D microstructural datasets are commonly used to define the geometrical domains in finite element modelling. This has proven a useful tool for understanding how complex material systems behave under applied stresses, temperatures and chemical conditions. However, imaging of materials is challenging number reasons, including limited field view, low resolution difficult sample preparation. Recently, machine learning method, SliceGAN, was developed statistically generate arbitrary size using...

Imaging is critical to the characterisation of materials. However, even with careful sample preparation and microscope calibration, imaging techniques are often prone defects unwanted artefacts. This particularly problematic for applications where micrograph be used simulation or feature analysis, as likely lead inaccurate results. Microstructural inpainting a method alleviate this problem by replacing occluded regions synthetic microstructure matching boundaries. In paper we introduce two...

Modelling the impact of a material's mesostructure on device level performance typically requires access to 3D image data containing all relevant information define geometry simulation domain. This must include sufficient contrast between phases distinguish each material, be high enough resolution capture key details, but also have large field-of-view representative material in general. It is rarely possible obtain with these properties from single imaging technique. In this paper, we...