La génération de formes naturelles a été le sujet de nombreuses recherches depuis plusieurs années. Plusieurs méthodes ont été proposées afin de générer des objets naturels et réalistes tels que des terrains, des plantes et arbres, des nuages, etc. Les modèles itératifs sont très connus dans ce domaine de recherche grâce à leur capacité à générer des formes rugueuses et complexes qui sont adaptées à la représentation d’objets naturels. L’inconvénient majeur de tels modèles est le manque de contrôle sur le résultat final. Ce dernier peut venir de la méthode de construction stochastique interdisant tout contrôle par définition. Pour les modèles dont la construction est déterministe, les paramètres de générations sont souvent non intuitifs et limitent ainsi le contrôle. Pour ces raisons un grand nombre de recherches ont port sur le problème du contrôle de ces modèles ainsi que sur la possibilité d’utiliser des modèles non-itératifs (esquisses, basés exemples, etc.). Bien souvent, le contrôle introduit par ces modèles est un contrôle global, c’est-à-dire sur la totalité de l’objet final et ne prend donc pas en compte les détails locaux de ce dernier. Dans notre travail, nous nous attaquons au problème du contrôle sur les formes naturelles en tenant compte du contrôle local. À cette fin, nous introduisons deux modèles différents. Le premier repose sur un formalisme itératif avec notion de détail qui se dicline en deux sous-familles, l’une basée sur les IFS et l’autre basée sur les surfaces de subdivision. Le deuxième modèle permet l’édition de caractéristiques d’un terrain sous forme de primitives vectorielles puis la génération du terrain par une méthode de diffusion guidée. Cette dernière fait l’objet d’une implémentation parallèle sur la carte graphique (GPU). / The generation of natural shapes has been the subject of much research for many years. Several methods have been proposed to generate realistic natural objects such as terrain, plants and trees, clouds, etc... Iterative models are well known in this field of research due to their ability to generate complex and rough shapes that are adapted to the representation of natural objects. The major drawback of such models is the lack of control over the final result. The latter can come from the stochastic construction method which prevents any control by definition. For models whose construction is deterministic, the parameters of generation are often non-intuitive and thus limit control. For these reasons many studieshave focused on the problem of controlling these models as well as the possibility of using non-iterative models (sketches, based on examples, etc). Often, control introduced by these models is a global control, on the whole final object and therefore does not include local details of this object. In our work, we focus in the problem of control over natural shapes, taking into account local control. To this end, we introduce two different models. The first is based on an iterative formalism with detail concept which is divided into two subfamilies, one based on IFS and the other one based on subdivisionsurfaces. The second model allows the editing of terrain features under a form of vectorial primitives which one used to generate the terrain by guided diffusion method. The latter is the subject of a parallel implementation on graphics card (GPU).

I denna rapport kommer vi att beskriva vår process för att utveckla ett spelprojekt som vårt examensarbete genom att vara med och tävla i Sveriges största spelutvecklartävling för studenter, Swedish Game Awards. Spelet handlar om personen Kurt, som har bestämt sig för att ge sig ut för att rädda efter sin kidnappade hund.   I denna rapport förklarar vi hur vi gick tillväga när vi fortsatte utvecklingen av ett spel samt vilka verktyg och metoder vi använde för att fullborda uppgiften, både genom de krav som ställdes från Swedish Game Awards och de krav vi bestämde själva. Vi beskriver de motgångar vi hade och hur vi lyckades lösa problemen. Slutligen kommer vi även redovisa vårt resultat och diskutera kring utvecklingen av spelet, samt dess utvecklingspotential för framtiden. / In this report we are going to describe the development process of our game that we have as our thesis. To be able to have this project as our thesis, we had to enter our game in the game development competition called “Swedish Game Awards”. This competition is only for students, giving them a place to present their game projects. The game is about a guy named Kurt, who decides to go and rescue his kidnapped dog.   We will explain how we went about developing the game and about the different tools and methods we used to complete the project. Both in terms of the demands that is needed to send our project as an entry to the Swedish Game Awards, and in terms of the requirements we decided on by ourselves. We will describe the adversity we had in our project and how we went about solving them. Finally we will also present and discuss our results from the development of the game, and its future potential for further development.

<p>Terrain generation is a convoluted and a popular topic in the VFX industry. Whether you are part of the film/TV or gaming industry, a terrain, is a highly nuanced feature that is usually present. Regardless of walking on a desert like terrain in the film, Blade Runner 2049 or fighting on different planets like in Avatar, 3D terrains is a major part of any digital media. The purpose of this thesis is about developing a workflow for large-scale terrains using complex data sets and utilizing this workflow to maintain a balance between the procedural content and the artistic input made especially for smaller companies which cannot afford an enhanced pipeline to deal with major technical complications. The workflow consists of two major elements, development of the tool used to optimize the workflow and the recording and maintaining of the efficiency in comparison to the older workflow. </p> <p> </p> <p> My research findings indicate that despite the increase in overall computational abilities, one of the many issues that are still present is generating a highly advanced terrain with the added benefits of the artists and users’ creative variations. Reducing the overall time to simulate and compute a highly realistic and detailed terrain is the main goal, thus this thesis will present a method to overcome the speed deficiency while keeping the details of the terrain present.</p>

This thesis explores the possibilities of creating landscapes through procedural means within the game engine Unreal Engine 5. The aim is to provide a flexible procedural landscape tool that doesn't limit the user and that is compatible with existing systems in the engine. The research questions focuses on comparison to other work regarding landscape generation and generation of procedural roads. The process to achieve this was done through extensive implementation adding modules that both builds upon and adds to the source code. The implementation was divided into five major components, which was noise generation for terrain, biotope interpolation, asset distribution, road generation and a user interface. Perlin noise, utilizing Fractal Brownian Motion were a vital part of generating terrain with varying features. For interpolation a modified version of Lowpass Gaussian filtering was implemented in order to blend biotope edges together. Asset distribution and road generation were implemented in a way that uses pseudo-randomness combined with heuristics. The user interface was done to tie everything together for testing. The results shows potential for assisting in procedural landscape creation with a large amount of freedom in customization. There is however flaws in some aspects, namely the interpolation methods suffer from clear visual artefacts. Whether it is suitable for professional standards remains to be fully proven objectively as the testing in this thesis work was limited.

The power of generative deep learning has increased very quickly in the past ten years and modern models are now able to generate human faces that are indistinguishable from real ones. This thesis project will investigate the uses and limitations of this technology by attempting to generate very specific data, images of golf holes. Generative adverserial networks, GANs, were used to solve this problem. Two different GAN models were chosen as candidates and these were trained on some different datasets that were extracted from the project provider Topgolf Sweden AB’s virtual golf game. This golf game contained data of many different types of golf holes from all over the world. The best performing model was Progressive Growing GAN, ProGAN, which works by iteratively increasing the size of the images until the desired size is reached. This model was able to produce results of very high quality and with large variety. To further investigate the quality of the results a survey was sent out to the employees of Topgolf Sweden AB. A survey that showed that it was difficult for the participants to correctly determine if a given image was real or had been generated by the model. These results further showed that the generated samples had a high quality. This thesis project also investigated how height data could be incorporated in the process. The results showed that the ProGAN model was able to generate height maps that capture the most important aspects of a golf hole. Furthermore, the overall results showed that the generative model had learned a good representation of the data’s underlying probability distribution. More work needs to be done before a model like the one presented here can be used to generate complete golf holes that can be used in a virtual golf game, but this project clearly shows that GANs are a worthwhile investment for this purpose. / Kraften i generativ djupinlärning har ökat snabbt under de senaste tio åren och moderna modeller kan generera bilder på människoansikten som är omöjliga att urskilja från riktiga ansikten. Detta examensarbete undersöker hur denna teknologi kan användas och vad det finns för begränsningar genom att försöka generera väldigt specifik data, bilder på golfhål. Generativa adversiella nätverk, GANs, användas för att lösa detta problem. Två modeller valdes som kandidater och dessa tränades på olika datasets som hade extraherats från projektleverantören Topgolf Sweden ABs virtuella golfspel. Detta golfspel innehöll data från en mängd olika typer av golfhål från hela världen. Modellen som presterade bäst var Progressive Growing GAN, ProGAN, som iterativt ökar storleken på bilderna tills den önskade storleken har nåtts. Denna modell lyckades skapa bilder av väldigt hög kvalitet och med stor variation. För att ytterligare undersöka kvaliten på resultaten så genomfördes en enkät. Enkäten skickades till anställda hos Topgolf Sweden AB. Svaren visade att det var svårt för deltagarna att urskilja äkta bilder från genererade bilder vilket ytterligare visade att de genererade bilderna hade hög kvalitet. Detta examensarbete undersökte också hur höjddata kunde integreras i processen. Resultaten av detta visade att ProGAN modellen kunde generera höjddata som innehöll de viktigaste delarna av ett golfhål. Dessutom så visade resultaten i helhet att den generativa modellen hade lärt sig en bra representation av träningsdatans underliggande sannolikhetsfördelning. Mer arbete krävs för att en liknande modell ska kunna generera kompletta golfhål som kan användas i ett virtuellt golfspel, men projektet visar att GANs är ett väldigt bra alternativ för att lyckas med det.

This Master s thesis proposes a non-assisted procedural method for 3D canyons scenes generation based on techniques of computer graphics, computer vision and graph search algorithm. In order to define all the features to be reproduced in our scenes, we have analyzed several images of real canyons and have categorized them in two canyon features models: a recursive and an ordinary one. The proposed approach manipulates a heightmap, created using Perlin noise, in order to imitate the geological features formation previously analyzed. Several parametrizations are used to guide and constraint the generation of terrains, canyons features, course of river, plain areas, soft slope regions, cliffs and plateaus. This work also uses the Mean Shift algorithm as mechanism of segmentation to define regions of interest. A binary mask, with plain areas, is defined based on a threshold operation by a given data set provided by the Mean Shift algorithm. Thereafter a connected-component labeling algorithm is executed using the previously binary mask. This algorithm finds all plains centroids. Right after that, the Dijkstra s algorithm is performed in order to connect all plain areas, creating a valid path between the centroids. The Dijkstra s algorithm is executed again to define the river s course. Finally, a Gaussian smoothing operation is applied to interpolate the soft slope regions. The combination of all those techniques produces as a result automatically generated feature-rich canyons. / Esta dissertação propõe um método procedural não assistido, baseado em técnicas de computação gráfica, visão computacional e busca em grafos, para a geração de cenários 3D de cânions com foco em jogos digitais. Para definir as características a serem reproduzidas, foram analisadas diversas imagens de cânions reais chegando-se em dois modelos, um comum e outro recursivo. A abordagem proposta manipula um reticulado gerado com ruído de Perlin, moldando assim as características inerentes a essa formação geológica. São levadas em conta as diversas parametrizações necessárias para permitir que o algoritmo construa cânions com curso de rio, áreas de planícies, regiões de encosta suave, estruturas de penhascos e, por fim, planaltos nas regiões mais altas. Para atingir o resultado final, o trabalho utiliza o algoritmo Mean Shift como mecanismo de segmentação, definindo dados e regiões de interesse. Munido dos dados do algoritmo de clusterizacao, é definido um limiar para a criação de uma máscara binária com a definição das planícies. Em um segundo momento, um algoritmo de rotulação de componentes conectados é executado, extraindo-se os centróides de cada planície. Por sua vez, o algoritmo de Dijkstra encaixa-se na definição de rotas que conectam estas planícies. O algoritmo de Dijkstra é, então, executado novamente, tendo por base uma função de custo de inclinação, para definir o curso do rio. Por fim, uma filtragem espacial baseada em um filtro Gaussiano é aplicada para interpolar as regiões de encostas de declive suave. A combinação dessas técnicas gera terrenos com grande variabilidade e com as características inerentes à formação geológica de cânions.

The field of this project is size restricted graphic intro. The paper deals with motivation creating such an intro and talks briefly about history. Main focus is put on general description of various principles common in demo development, techiques used achieving the practical assignment are provided with more detailed description.