Adhesives are essential components of everyday life and have been for thousands of years. The history of adhesives begins with a stone tool covered in birch tar found at Campitello Quarry, Italy, dating to around 200,000 years ago. This find demonstrated the use of adhesives by Neanderthals and their ability to manufacture materials through transformative processes. In Southern Africa, modern humans have been producing compound adhesives by mixing plant and mineral materials since at least 70,000 years ago, tailoring adhesives to different environments and uses. In recent years, the number of identified prehistoric adhesives has grown, and adhesive technology has become a proxy for discussing technological complexity across different hominin species. However, to fully evaluate and compare the adhesive technology of other human groups across space and time, more research is needed on the selection of adhesive materials, adhesive recipes, and the context in which adhesives and their tools were used.

With this dissertation, I contribute to enlarging the sample of identified prehistoric adhesives by analysing tools with adhesive residues from Steenbokfontein Cave (South Africa, Later Stone Age), Morín Cave (Spain, Middle-Upper Palaeolithic), and the Dutch North Sea (the Netherlands, Mesolithic). I employ a multi-analytical approach encompassing optical microscopy and experiments with the integration of data from chemical analysis of residues. The results of the analysis, combined with data from available literature, provide insights into several questions that enhance the debate on the technological complexity of Neanderthals and modern humans. What materials and additives were used by Neanderthals and modern humans to manufacture adhesives? Was there a difference in adhesive recipes depending on the context of use of the hafted tools? Was there a preference for hafting specific tools? Is there a difference between Neanderthal and modern human adhesive technologies in terms of raw materials exploited, use of additives, and context of use? Does adhesive technology reflect differences in technological complexity between Neanderthals and modern humans?

All the analysed adhesives were used by Neanderthals and modern humans to fasten their tools to organic handles. Adhesive residues have been identified on stone and organic projectile points, as well as on ‘common tools’ used in domestic tasks, strongly suggesting that adhesives were integrated into the domestic economy of Neanderthals and modern humans. No relevant differences in the use of adhesives were observed depending on the tools’ raw materials or functions.

Adhesives were mostly produced from natural resources available in the surrounding environment. At Steenbokfontein Cave, South Africa, adhesives were made using the resin or tar of conifer trees, specifically Podocarpus or Widdringtonia, both of which were available near the Cave and mixed with (mineral) additives. Similarly, at Morín Cave in Spain, the resin from a tree of the genus Juniperus, largely available in the environment, was likely used. However, there is evidence that some adhesives were selected over others equally available for their material properties. For instance, birch bark tar was preferred over pine resin for hafting bone points at the Dutch North Sea, a trend seen at many other Mesolithic sites. Furthermore, increasing evidence suggests that Neanderthals used additives, primarily iron oxides, to alter the material properties of their adhesives, similar to contemporaneous modern humans in Africa. This reflects Neanderthals' and modern humans' understanding of available natural resources, their distinct material properties, and the effects of their combinations.

Adhesive technology requires good knowledge of natural resources and their material properties, control of fire, enlarged cognitive functions, and forms of cultural transmission and social learning, qualifying it as a complex technology. The examination of adhesive remains in this thesis demonstrates that Neanderthals and modern humans share considerable technological parallels, highlighting Neanderthal technological sophistication. How Neanderthals selected, transformed, and employed adhesives suggests analogous procedures and reasoning to modern humans. Consequently, these insights likely reflect that Neanderthals had comparable cognitive and technological skills to anatomically modern humans.

This dissertation develops intrinsic approaches to learning and computing on curved surfaces. Specifically, we work on three tasks: analyzing 3D shapes using convolutional neural networks (CNNs), solving linear systems on curved surfaces, and recovering appearance properties from curved surfaces using multi-view capture. We argue that we can find more efficient and better performing algorithms for these tasks by using intrinsic geometry.

Chapter two and three consider CNNs on curved surfaces. We would like to find patterns with meaningful directional information, such as edges or corners.
On images, it is straightforward to define a convolution operator that encodes directional information, as the pixel grid provides a global reference for directions. Such a global coordinate system is not available for curved surfaces. Chapter two presents Harmonic Surface Networks. We apply a 2D kernel to the surface by using local coordinate systems. These local coordinate systems could be rotated in any direction around the normal, which is a problem for consistent pattern recognition. We overcome this ambiguity by computing complex-valued, rotation-equivariant features and transporting these features between coordinate systems with parallel transport along shortest geodesics.

Chapter three presents DeltaConv. DeltaConv is a convolution operator based on geometric operators from vector calculus, such as the Laplacian. A benefit of the Laplacian is that it is invariant to local coordinate systems. This solves the problem of a missing global coordinate system. However, the Laplacian operator is also isotropic. That means it cannot pick up on directional information. DeltaConv constructs anisotropic operators by splitting the Laplacian into gradient and divergence and applying a non-linearity in between. The resulting convolution operators are demonstrated on learning tasks for point clouds and achieve state-of-the-art results with a relatively simple architecture.

Chapter four considers solving linear systems on curved surfaces. This is relevant for many applications in geometry processing: smoothing data, simulating or animating 3D shapes, or machine learning on surfaces. A common way to solve large systems on grid-based data is a multigrid method. Multigrid methods require a hierarchy of grids and the operators that map between the levels in the hierarchy. We show that these components can be defined for curved surfaces with irregularly spaced samples using a hierarchy of graph Voronoi diagrams. The resulting approach, Gravo Multigrid, achieves solving times comparable to the state-of-the-art, while taking an order of magnitude less time for pre-processing: from minutes to seconds for meshes with over a million vertices.

Chapter five demonstrates the use of intrinsic geometry in the setting of appearance modeling, specifically capturing spatially-varying bidirectional reflectance distribution functions (SVBRDF). A low-cost setup to recover SVBRDFs is to capture photographs from multiple viewpoints. A challenge here, is that some reflectance behavior only shows up under certain viewing positions and lighting conditions, which means that we might not be able to tell one material type from another. We frame this as a question of (un)certainty: how certain are we, based on the input data? We build on previous work that shows that the reflection function can be modeled as a convolution of the BRDF with the incoming light. We propose improvements to the convolution model and develop algorithms for uncertainty analysis fully contained in the frequency domain. The result is a fast and uncertainty-aware SVBRDF recovery on curved surfaces.

The goal of this project was to simulate the colour change over time of Girl with a Pearl Earring from Johannes Vermeer. Using art history knowledge and computational methods this was proven to be possible. The combination of these methods is unique and promising.

As a first for the Girl, the paints that Vermeer used have been reconstructed and artificially aged by UV light at AkzoNobel. Their colour change was then measured and turned out to be as expected. The mechanisms of indigo, yellow and red lake fading seems plausible. A start in reconstructing the painting has been made by digitally applying these colour changes on the original painting.

The digital reconstruction showed that the contrast increased and the Girl is more lively and looks ready to step out of the painting, there is a clear three dimensional effect. The colours are more defined instead of faded and patchy. More contrast can be seen, which is for example shown in the white collar. It also makes the edges between the face and background sharper. Overall, the painting shows more Houding and seems to be more like Vermeer probably intended it to be. A reconstruction made with BRDF should be more scientifically accurate than the reconstruction presented here.

Important points for future research are more opaque samples, which are created in more batches. Another point is that the measured light intensities should not be in lux and to be quick in the BRDF measurements after UV aging. Finally, it would be interesting to reconstruct other paintings, from Vermeer or from another painter.

Hyperspectral techniques have found application for the investigation of objects and samples in many scientific fields. Data evaluation approaches commonly either process them as stacks of randomly ordered spectra or as flat images. This thesis aims at combining both, lateral and spectral, aspects of hyperspectral data during data evaluation. Different approaches, such as lateral distance and neighbor pixel augmentation, will be explored with modern factorization techniques, such as t-stochastic neighbor embedding and self-organizing maps, and supported by artificial neural networks. The routines developed are applied on selected X-ray fluorescence data (XRF) sets from the field of materials science and cultural heritage. The altered augmented data set concept consists in augmenting the spectrum of a central pixel with the mean spectrum of its eight neighboring pixels. This new method is applied to the data set of a bi-modal Ti-6Al-6V-2Sn alloy. Clustering the optimized augmented data set with t-SNE reveals the existence of four components: phase with strong titanium signal, phase with strong vanadium signal, the first phase on top of the second phase and the second phase on top of the first phase. The later was not identified by clustering with t-SNE the data set augmented with all the spectra of its eight neighboring pixels. Clustering the XRF data set of a 13th century B.C. Egyptian mural painting with Self Organising Map (SOM) indicates areas with different thicknesses of copper and iron containing pigments. Clustering with Fast interpolation-based t-SNE (FIt-SNE) also reveals information about the painting sequence and the composition of a mixture of pigments. The pigments are not separable with a simple observation of the XRF elemental distribution images. Moreover, FIt-SNE dramatically accelerates t-SNE and can provide well-separated clusters. These properties are especially useful for large data sets, containing mixture of pigments. Finally, I introduce an optimised Artificial Neural Network (ANN) for training the photograph of a painting with the elemental distribution images, obtained by its XRF data set. The elemental distribution images are upscaled with Laplacian Pyramid Super-Resolution Network. In this way, we can identify with more accuracy the exact location of the damages and retouches on a 17th century A.C. easel painting (portrait of Hortense Manchini), since the result image is of high resolution.

The scientific analysis of historical paintings has been traditionally restricted to the analysis of paint cross-section samples. This invasive method provides extensive information but is inherently limited in scope due to the extreme heterogeneity of paintings. In the last decade, non-invasive spot analyses and spectral imaging methods have become increasingly widespread in cultural heritage science. Two of these methods are macroscopic X-ray fluorescence imaging spectroscopy (MA-XRF) and reflectance imaging spectroscopy (RIS). These methods allow for 2D-scanning the entire surface of a painting and provide complementary information on elemental and molecular composition and distribution of the paint. However, these methods are often used only for qualitative analysis of the paint based on relative distribution maps, revealing only limited information about the paint layer stratigraphy. This thesis is an exploration of a combined approach for quantitative analysis of paint composition and layer stratigraphy using MA-XRF and RIS. The research used a set of specially prepared paint samples of mixtures and multiple layer applications based on historically relevant pigments which were scanned using MA-XRF and RIS in the visible and Near IR range (400-2500 nm). The spectral data acquired were processed and analyzed in a variety of ways, including Non-negative Matrix Factorization, Non-Linear Least Square Fitting, among other methods; in an attempt to gather quantitative compositional and stratigraphic data. In these trials, using characteristic reflectance features in the visible range together with a comparison of highly and lightly absorbed X-ray fluorescence lines allowed for the identification and quantification of surface specific compounds related to the top paint layer. Further comparison of fluorescence lines and absorption features in the Near IR range provided a potential avenue for quantification of subsurface paint layers. The results confirm that the combination of these methods allows to reveal the paint stratigraphy. The project provides samples and data sets which may serve as the basis for the development of a robust algorithm to address this issue in the future.

A novel combination of analytical techniques for measuring optical properties of glaze and paint as input for computer modelling is used to investigate the influence of the green glaze layer present on top of the black underpainting in the background of Girl with a Pearl Earring. The absorption spectra of paint layers from paint reconstructions and samples of the original painting were determined using micro spectrophotometry. Additionally, Bidirectional Reflectance Distribution Function (BRDF) measurements were done on paint reconstructions to determine the reflection distribution from multiple illumination and observation angles. Already existing height and layer thickness information was compiled in a digital 3D model of the painting which, in combination with the optical properties, was used as a base for rendering using the Mitsuba render engine. The experimental results clearly show that the green glaze layer under specific angles darkens the appearance of the black paint onto which it is applied. The computer simulation shows promising results for digitally recreating the painting.

The appearance of a painting cannot solely be described by the depiction that it presents to the viewer. When viewing the artifact in real life, we find that the painted surface is in effect a three-dimensional landscape of paint. Paintings, “moveable, largely two-dimensional images created for the primary purpose of providing a visual experience”,1 can be created using a vast variety of materials on a range of supports. They are commonly built up as a complex stratigraphy of layers, generally consisting of a support, ground layer(s), one or multiple layers of (semi-) transparent paints, and in many cases a protective varnish layer. The current appearance of a painting is determined by the way a painter used and applied the materials, but also effects of aging, conservation and restoration treatments, which all continue to influence the physical state of a painting. Historically, cultural heritage (CH) reproductions were hand-crafted, and created for instance to disseminate or replace artworks, or to train in the skill of their creation. Also modern reproductions — or facsimiles — are still large hand-crafted, and for instance serve to provide access to (fragile) artworks or even complete (CH) sites, or to recreate their original appearance. Alternatively, reconstructions might reside only in the virtual domain. The continued development of digital imaging and digital fabrication technology (i.e. 3D printing) provides new opportunities for appearance reproduction, also suitable for application in the CH domain. If we want to replicate material appearance, we need to understand how (the appearance of) material is perceived. A material is, however, not observed directly, but has to be lighted, and via the light that is scattered by the material humans can perceive it. Appearance is therefore the light-material-confounded proximal stimulus for the human visual system (HVS). Even though we see, recognize and interact with a vast number of materials every day, and can effortlessly distinguish between them, it turns out that the perceptual mechanisms that underlie this, are still quite poorly understood, including linking individual appearance attributes to measurable and fabrication parameters. One of the consequences of this is that an integrated approach to (total) appearance reproduction, including color, topography/ texture/shape, gloss and transparency/translucency, is still lacking.

It is assumed that 17th century paintings were painted by a very systematic approach which involved the use of very specific palettes. For every pictorial unit, such a skin, a very specific palette was created. In this case the palette would then only be used to paint the skin and nothing else. The result is that for each pictorial unit a separate palette is used. This assumption can be verified by characterizing the all the paint mixtures present in a painting. Paint derives it colour from the pigments present within them. Many pigments used in the 17^th century contain characteristic heavy elements. The usage macro X-ray fluorescence (XRF) scanning is gaining popularity as a non-destructive method for determining elements present in a paintings. By the creation of elemental maps the pigments can be identified and localized in a painting. These maps already allow for visualizing many interesting features such as retouches, inpaintings and even hidden layers behind a painting. However the usage of the XRF data have never been used for identifying and visualizing the palettes used while painting.
In this work it is attempted to identify and visualize the palettes used in the Man with the red baret by Rembrandt van Rijn, by an in-depth analysis of raw XRF data. The raw data was model fitted using a simulated annealing algorithm to retrieve spectral line intensities in each point of the painting. The fitted spectral intensities was then visualized using t-distributed stochastic neighbour embedding (t-SNE). With help of the t-SNE visualisation the painting is divided into 24 sections. The suggestion is that these 24 sections each contain a different and unique paint mixture. This suggestion is further investigated by the use of the two-sample χ²-test between the sections’ spectral composition. The combination of model fitting alongside with t-SNE on a XRF spectra can indeed visualize different paint mixtures and give strong suggestion of the different palettes used in Man with the red baret.

Scientific investigation of paintings has been facilitated by the development of advanced non-de\-struc\-tive imaging methods. Characterization of painting stratigraphy traditionally requires extraction of small paint samples, thereby limiting its use to a few locations on a painting due to its destructive nature. Alternatively, non-destructive analysis of paint layer stratigraphy and structure across an entire painting often requires highly specialized and costly equipment, and/or the transport of priceless artworks. In addition, most methods are also typically limited to a localized point analysis.

This document proposes an alternative method for the substructure examination of paintings using a mobile macro-XRF spectrometer and a stereographic approach with reduced step sizes. This is coupled with a novel data analysis method which will enable a global study of the topographical features of hidden paint layers. As a prototype to test the feasibility of the method, we utilized a two-layer test sample consisting of pastose bone black pigment on pastose lead white, with an aluminum substrate. High resolution 3D optical microscopy was utilized to establish a ground truth for the thicknesses of both paint layers.

Through successful registration of MA-XRF scans obtained with varying detector geometries into a single hyrbid image, we were able to find a strong correlation between the quanitative height data obtained with optical microscopy and our hybrid XRF image. Our findings indicate that utilizing this approach for visualizing hidden paint layer topographies proves to be very promising. Coupled with novel data fusion algorithms and visualization techniques, additional insight about painterly technique can thus be gained by utilizing existing MA-XRF scanners to scan a painting in multiple orientations.

Rembrandt was a unique painter of his era, and one of his most outstanding pieces is The Jewish Bride. He mastered the art of manipulating pigments to create unique paint mixtures and textures. This can be seen in his painting, The Jewish Bride.
In the background of The Jewish Bride, Rembrandt mixed five pigments together: bone black, red earth, smalt (a blue pigment), yellow earth and yellow lake. Smalt and yellow lake are unstable pigments in oil and their colour fades over time. Due to the degradation process of smalt, the oil matrix and the top layer have both changed its appearance into a transparent monochrome brown. The colour loss and change of colour affect the pictorial impression created by the painting.
The goal of this thesis is to digital colour reconstruct the background of The Jewish Bride to show how it might have looked when non-degraded smalt and yellow lake were present.
In this work, the painting is analysed with different micro and macro techniques: microscopy, SEM-EDX, Macro-XRF, and hyperspectral imaging. The outcomes of microscopy, SEM-EDX and Macro-XRF are combined to show the distribution and to quantify the amount of smalt and yellow lake in the background.
Knowing the distribution of smalt and yellow lake in the painting made it possible to digital mix the colours back to their primary appearance. To achieve this, the hyperspectral data were used, together with the Kubelka-Munk theory. This was later transferred to RGB values, which results in a digital colour reconstruction.
The digital colour reconstructions show a trend of what the painting might have looked like, at the time Rembrandt painted it.

The different phases in the life of archaeological objects can be described by artefact biography. This dissertation defines an updated version: artefact biography 2.0, and the life phases of Early Iron Age bronze studs from Oss-Zevenbergen, the Netherlands, are elaborated. Throughout the thesis, the research revolves around information value.

The bronzes have been studied from an interdisciplinary point of view, which has allowed the extraction of information that is also applicable to other corroded bronzes in general. It is argued that:
- corrosion products and inclusions may reflect original microstructure

- corrosion inhibitor BTAH binds to Sn and SnO2
- technical investigations before conservation increase the information value of an artefact.