An abstract is usually brief, it must do almost as much work as Time travel is an old dream of mankind that is fueled by fascination and curiosity. Of course, such a journey through time goes far beyond our physical possibilities today. Historical science largely recapitulates the past using text, while the creative industry reproduces it as a more or less well-researched fiction. The Time Machine Initiative has now taken on the task of digitizing and processing the cultural heritage on a large scale in order to create new virtual accesses to the past, which - taking into account the fragmentary tradition - come close to a journey through time. In a large-scale interdisciplinary and trans-European research project, a kind of edition of European history is to be created which, as a data-saturated reconstruction, can create a new form of comprehensibility and experience. The time machine would therefore be a virtual research environment. In order to reach this ambitious goal, digitization and automated analysis of history and art has to be taken one step at a time. In the presentation, we shortly describe the project and present concrete research results that have been obtained in this direction ranging from book CT, i.e., the scanning of a whole book in a single scan, over writer and font identification up to first results on automated art analysis and understanding.

Historic clusters of heritage buildings comprise the core of a great number of European cities and represent the fabric based on which today's municipalities have developed historically. The sustainable development of these environments is often threatened by urbanization, gentrification or depopulation phenomena. These urban environments should not be studied and analysed as static formations disconnected from the contemporary fabric of a city, but rather as an assemblage of tangible and intangible assets subjected to dynamic pressures of economic, environmental, and social activities. The value of the historic built environment for local communities, as a tangible result of the cultural heritage of a place, does not only lie in preserving a continuity with past societies, but it can become important in achieving more resilient futures for the city [1]. The cross-disciplinary nature of the pressing challenges posed by said phenomena requires the development of novel data-driven methods [2] for the re-use, regeneration and safeguarding of neglected areas of our cities' existing building stock. Digitisation of the construction industry [3] and urban data analytics [4] offer new opportunities for historic cities that undergo transformations. The presentation will discuss about the methodological and technical framework required for the creation of a platform that will function as a time machine of our cities in the future. A time machine that does not aim only at representing how our cities used to be in the past, but rather one that curates and stores current transformations of our built environment, with the objective to enable dynamic observation of the existing building stock at neighborhood scale in present and future times. In this context, the presentation will be occupied with the significance of bringing the building scale (architectural) data together with neighbourhood scale (environmental) data in the same digital environment to enable deeper and cross-disciplinary insights of built heritage assets' conditions. This data-driven study is enabled by the use of Building Information Modelling (BIM) tools for the common management of multi-scale and multi-discipline datasets generated by the 3D documentation, non-destructive testing and metadata integration of conservation state analyses and historic architecture information of building assets. In this context, the presentation will be occupied with the significance of bringing the building scale (architectural) data together with neighbourhood scale (environmental) data in the same digital environment to enable deeper and cross-disciplinary insights of built heritage assets' conditions. This data-driven study is enabled by the use of Building Information Modelling (BIM) tools for the common management of multi-scale and multi-discipline datasets generated by the 3D documentation, non-destructive testing and metadata integration of conservation state analyses and historic architecture information of building assets. Finally the presentation will offer a description of the requirements for integrating these datasets in online repositories for the open access of the public and relevant stakeholders to spatial data analytics that can be used for territorial planning, energy monitoring, educational purposes and smart historic city applications [5]. This research responds to the need for storing, accessing, analysing, and updating heterogeneous data of heritage buildings, which currently, are found in unstructured data repositories of in scattered, inaccessible databases.

Silver tarnish manifests by changes in the optical properties of the material. Documenting these changes creates many challenges for imaging techniques. This paper proposes a methodological approach based on processing Reflectance Transformation Imaging (RTI) data for tracking multi-temporal changes on such surfaces. Through the statistical analysis of the surface's angular reflectance, information related to the appearance attributes can be evaluated and visualized by maps. Thus, this paper explores the global surface change of the reflectance response of silver tarnishing as a function of time. A qualitative and semi-quantitative evaluation is based on multivariate distance measurements at different time intervals. The results are compared to surface change evaluation by visual inspection, photographic documentation, and colourimetry, practices traditionally used in conservation documentation to monitor surface changes over time. The outcome of this research illustrates the possibilities of RTI data analysis as a tool for accurate multi-temporal documentation of the optical properties changes on specular surfaces.

Carpet attribution is an important task for studying the carpets and textiles, and more generally the history and culture of the communities producing these carpets. However, this is not an easy task, often relying on experts' subjective opinion or complex chemical or radiographical analysis, often not available to many practitioners. In this work, building on the success of applying machine learning and artificial intelligence methods in different fields, we present another, data-driven approach for carpet attribution. Based on a large dataset of Azerbaijani flat woven carpets we have developed a novel machine learning based data-driven carpet attribution system, which successfully determines their types, schools and weaving century, achieving up to 98% accuracy of the attribution.

Painter attribution is based on a variety of factors, oftentimes deeply buried in the details such as the brushstrokes of the ears or the eyes, which a painter might paint in a specific way. To get to this details, the images have to be examined carefully and intensively. Our work is focused on this phenomenon of painter attribution, investigating those details using supervised machine learning methods for image recognition that rely on a set representation. In this paper however, we are going to focus on one step of our work specifically: The annotation process. With such a focus on details, a dense and detailed, but also transparent annotation of the images is necessary. On one hand this is essential for our research, on the other hand however, it is very time consuming and requires a lot of human resources. Therefore we developed an ontology for the annotation of the images and a semi-automated workflow with object detection component using YOLOv3 and closely tied to our ontology. This way we were able to automate our processes as efficiently as possible while maintaining the complexity of our annotations.

The nature of archaeological research implies documenting and recording the remains or structures uncovered in the most precise and objective way possible. Archaeologists use digital tools precisely because they meet the challenges of their discipline. The creation of digital twins thus actively contributes to the study, protection and dissemination of archaeological heritage. From the study of a territory to the analysis of a trace left by a tool, archaeology reasons at different scales and combines analyses and multi-scalar approaches. This represents a challenge in itself. The development of an open-source photometric stereo suite makes it possible to respond in part to this work by taking into account the great diversity of archaeological remains while producing models that combine volume accuracy and color reliability.

Identification of changes from cultural heritage (CH) surfaces incor- porates several factors like noise from the surface, error from the acquisition system, and alignment of the two phases of information in a one-time frame. In the post-processing pipeline for change iden- tification, the alignment always generates a bias in calculating the changes. This work proposes a pipeline for processing the surface normal calculated from a simulated Reflectance Transformation Imaging (RTI) acquisition. In this work, we have proposed a normal distribution analysis of the neighboring pixels to give more confi- dence to the change detection method. To claim the ground truth of the segmentation method based on a normal distribution, we have decided to work on the simulated RTI acquisitions. This will help us eliminate the mentioned errors and noises and check their validity. We have considered a visual inspection of the normal distribution of the neighboring pixels and set several parameters to group the several behaviors of the surface changes. From the segmentation, a semi-quantitative change calculation is also possible based on the segmented pixel count in each surface stage. The considered parameters were normalized to make the method independent of the acquisition parameters such as camera and light positions and magnification.