<p dir="ltr">With the recent virtualization of our everyday lives and the development of intelligent AI technology, realistic virtual agents are becoming a useful tool for research, education, and entertainment. With virtual agents customized content can be created and individualized for specific users. However, virtual agents are only effective tools if they can form a connection with the individual. To form a connection the virtual agent must be believable.</p><p dir="ltr">The literature review evaluates topics of virtual agent believability, appeal, and naturalness and how they relate to asymmetry in facial expression animation. The literature suggests that asymmetries can affect the perception of virtual agents. Additionally, it suggests that emulating human behavior is beneficial to increasing perception of believability, appeal, and naturalness.</p><p dir="ltr">In this study we evaluated the effects of facial expression asymmetry on the believability, appeal, and naturalness of virtual agents. To do this we ran an online perception study with students at Purdue University. We found, in brief, that facial expression asymmetries do have significant effect on the believability, appeal, and naturalness of a virtual agent compared to animations that do not include facial expression asymmetries.</p>

This paper aims to investigate the effect of AI-generated content (AIGC) when it starts to be applied in game development. AIGC in games refers to the generation of game content through artificial intelligence, a concept that has recently recieved a high level of attention due to the latest rapid developments in artificial intelligence, and in traditional research, AIGC can be categorized as an advanced approach to Procedural Content Generation (PCG), i.e., Deep Learning Method. Procedural Content Generation is the creation of game content through algorithms with limited or indirect user input. Its traditional approach has been widely used in games. Recently, however, the AIGC method has also started to be used by a large number of game companies, and its impact has exceeded expectations. A questionnaire survey of 40 game developers revealed a general interest in AIGC but also concerns. Further interviews explored the use of AIGC in game development and some of the problems it has encountered and predicted future trends in its development. The result of this study provide guidance on whether and how AIGC needs to be used in future game development.

Digital painting is the field of software designed to provide artists a virtual medium to emulate the experience and results of physical drawing. Several hardware and software components come together to form a whole workflow, ranging from the physical input devices, to the stroking process, to the texture content authorship. This thesis explores an artist-friendly approach to synthesize the textures that give life to digital brush strokes. Most painting software provides a limited library of predefined brush textures. They aim to offer styles approximating physical media like paintbrushes, pencils, markers, and airbrushes. Often these are static bitmap textures that are stamped onto the canvas at repeating intervals, causing discernible repetition artifacts. When more variety is desired, artists often download commercially available brush packs that expand the library of styles. However, included and supplemental brush packs are not easily artist-customizable. In recent years, a separate field of digital art tooling has seen the popular growth of node-based procedural content generation. 3D models, shaders, and materials are commonly authored by artists using functions that can be linked together in a visual programming environment called a node graph. In this work, the feasibility is tested of using a node graph to procedurally generate highly customizable brush textures. The system synthesizes textures that adapt to parameters like pen pressure and stretch along the full length of each brush stroke instead of stamping repetitively. The result is a more flexible and artist-friendly way to define, share, and tweak brush textures used in digital painting.

I denna uppsats kommer vi att presentera en modell vars syfte är att generera beteende-mönster för rollfigurer i digitala spel. Spelgenren rollspel (eng. Role-playing games) placerarspelaren i en värld fylld av fantastiska monster och modiga hjältar. I ett sådant spel ärde goda karaktärerna minst lika viktiga som de ondskefulla varelser som spelaren kämparmot. Hur ser livet ut för en spelkaraktär när den inte hjälper spelaren på dess äventyr? Dekanske lever som fiskare på havet eller som bönder ute på fältet. Mer troligt är att de bori en by tillsammans med en massa andra spännande karaktärer. Vi undersöker hur sådanakaraktärers vardag ser ut och tittar på en teknik som används för att skapa deras beteen-demönster, Cyclic Scheduling. Tekniken innebär att utvecklare skapar scheman som styrrollfigurernas beteende. Dessa scheman måste skapas i förväg under spelets utveckling ochkräver i stora spel många arbetstimmar för att utvecklas. Modellen vi presenterar i dennauppsats använder tekniken Smart Terrain för att automatiskt generera beteendemönsteroch kan användas för att minska utvecklingstid. Vi diskuterar hur modellen kan användas iföränderliga spelvärldar där utvecklare inte i förhand vet hur spelens omgivningar kommeratt se ut. / In this thesis we present a model for the generation of behaviour patterns for charactersin digital games. In the genre Role-playing games the player is placed in a world filledwith fantastic monsters and brave heroes. In this kind of game the good characters areas important as the evil creatures the player must fight against. What kind of life doesa game character have when not helping the player on adventures? Maybe they live asfishermen on the sea or as farmers in the fields. More likely they live in villages amongstother exciting game characters. We examine what these characters’ daily routines looks likeand look at a technique used for creating their behaviour patterns, Cyclic Scheduling. Thetechnique is used by developers to create schedules that control the behaviour of charactersin games. These schedules have to be created during the game development process andfor bigger games this consumes a lot of time. The model we presents in this thesis uses thetechnique Smart Terrain to automatically generate behaviour patterns thereby reducingthe development time. We discuss how the model can be used in dynamic game worldswhere the developers are unaware of potential changes in the game world.

I denna uppsats presenteras ett verktyg för att assistera skapandet av spelbanor i dungeon-miljö. Skapandet av sådana banor är ofta resurskrävande under produktionen och har fokus på design. För att behålla designaspekten men underlätta arbetet är verktyget framtaget för att med tillgängliga parametrar producera grottgångar för sådana banor. Vi undersöker med användartester hur verktyget kan användas för att effektivisera en känd metod för att skapa spelbanor, men samtidigt skapa den kvalité som eftertraktas. Med detta avser vi inte bara att ta fram ett effektivt verktyg, utan även att demonstrera en metod för att använda procedurell generering av spelinnehåll för ett nytt ändamål inom speldesign. / In this article a tool for assisting the creation of game levels in a dungeon environment is presented. Creating such game levels often requires a large amount of resources during a game production and has focus on design. To keep the aspect of design but ease work, the tool is created to produce caverns for such game levels with accessible parameters. We examine with user tests how the tool can be used to make an existing method for creation of game levels more effective. But at the same time producing the quality that is coveted. With this we do not only hope to produce an effective tool, but also to demonstrate a method for using procedural generation for a new purpose in game design.

Interactive evolution is explored and applied for the automatic generation of 8-bit sound effects (SFX). Procedurally generating a type of content can result in greater accessibility, cut development costs, and more. However, a natural problem that follows this approach is user fatigue. An 8-bit SFX generator is developed, tested, and then evaluated to understand its capabilities, usability, and effectiveness of several applicable solutions for reducing user fatigue. Results indicate that the software is intuitive to learn and use while providing a decent variety to the generated content with a probable feeling of progression. The implemented solutions for abating user fatigue show promise towards making the software practically viable. However, there are still areas in the developed artifact that suggest for further study.

The present research proposes two procedural content generation approaches for missions and play space in a game with dungeons, and a game prototype developed as a proof of concept for the feasibility of the algorithm and as a test framework for user-based experiments. The first approach will define missions by generating dungeon maps together with locked doors and keys through the use of an evolutionary algorithm. The second approach will populate the generated dungeon space by filling the content of dungeon rooms using space-filling curves and cellular automata algorithms. The evolutionary algorithm evolves tree structures encoding information about the dungeon. The goal is to converge the generated dungeons as close as possible to satisfy the set of parameter inputs provided by a game designer. The dungeon holds information about the amount of rooms, the connections between them and their position in a 2D map. There are also relevant semantic information in the content for the generation of narrative, which presents itself by the placement of unique pairs of keys and locks throughout it. Thus, a feasible way to finish the dungeon can be procedurally generated. The content of rooms are generated using space-filling curves algorithms such as Hilbert and Moore curves as well as Conways Game of Life Cellular Automata. Computational results report that the evolutionary algorithm provides dungeons with up to a 100 rooms very close to the desired ones for a range of different parameter inputs. The early validation tests with humans show no statistical difference between levels procedurally generated and those made by humans. Further user-centred validation tests with the game prototype show the algorithm-generated levels are perceived as equally or even more human-made than their human-authored counterparts, as well as funnier and more difficult. Thus, the research aims to generate gameplay elements combining different algorithms for a single solution, which could be easily adapted to a range of different games. / O projeto propõe duas abordagens de geração de conteúdo para missões e mapas em um jogo com calabouços, e um protótipo de jogo desenvolvido como prova de conceito da factibilidade do algoritmo e também como plataforma de testes para experimentos com usuários. A primeira abordagem define missões ao gerar mapas de calabouço em conjunto com chaves e portas trancadas através do uso de um algoritmo evolutivo. Já a segunda abordagem irá popular o espaço do calabouço criado ao preencher o conteúdo de suas salas usando algoritmos de curvas de preenchimento de espaço e autômatos celulares. O algoritmo evolutivo evolui uma estrutura em árvore que codifica informações sobre o calabouço. O objetivo é convergir os calabouços gerados para se aproximar ao máximo da configuração de entrada fornecida por um designer de jogos. O calabouço possui informação sobre as salas, como a quantidade das mesmas, as conexões entre elas e suas posições em um mapa 2D. Também contam com informações semânticas relevantes para a geração de narrativa no calabouço, que são o posicionamento de pares únicos de chaves e fechaduras através do calabouço. Portanto, uma maneira factível para o término do mesmo pode ser gerada proceduralmente. O conteúdo das salas é gerado usando curvas de preenchiment de espaço como as de Hilbert e Moore, além do autômato celular do Jogo da Vida de Conway. Resultados computacionais demonstram que o algoritmo evolutivo é capaz de criar calabouços com propriedades muito próximas às desejadas para uma grande variedade de entradas para calabouços com até 100 salas. Resultados preliminares de validação com humanos mostraram não haver diferença estatística entre os níveis gerados proceduralmente daqueles feitos por humanos. Testes posteriores de validação centrados em humanos com o protótipo de jogo mostram que os níveis gerados pelo algoritmo são percebidos como feitos por humanos de maneira semelhante ou até em maior grau do que suas contrapartidas geradas de fato por humanos, assim como são percebidos como mais divertidos e difíceis do que estas. Portanto, esta pesquisa objetiva gerar elementos de jogabilidade combinando diferentes algoritmos em uma única solução, que pode ser facilmente adaptada para uma variedade de jogos diferentes.

The present research proposes two procedural content generation approaches for missions and play space in a game with dungeons, and a game prototype developed as a proof of concept for the feasibility of the algorithm and as a test framework for user-based experiments. The first approach will define missions by generating dungeon maps together with locked doors and keys through the use of an evolutionary algorithm. The second approach will populate the generated dungeon space by filling the content of dungeon rooms using space-filling curves and cellular automata algorithms. The evolutionary algorithm evolves tree structures encoding information about the dungeon. The goal is to converge the generated dungeons as close as possible to satisfy the set of parameter inputs provided by a game designer. The dungeon holds information about the amount of rooms, the connections between them and their position in a 2D map. There are also relevant semantic information in the content for the generation of narrative, which presents itself by the placement of unique pairs of keys and locks throughout it. Thus, a feasible way to finish the dungeon can be procedurally generated. The content of rooms are generated using space-filling curves algorithms such as Hilbert and Moore curves as well as Conways Game of Life Cellular Automata. Computational results report that the evolutionary algorithm provides dungeons with up to a 100 rooms very close to the desired ones for a range of different parameter inputs. The early validation tests with humans show no statistical difference between levels procedurally generated and those made by humans. Further user-centred validation tests with the game prototype show the algorithm-generated levels are perceived as equally or even more human-made than their human-authored counterparts, as well as funnier and more difficult. Thus, the research aims to generate gameplay elements combining different algorithms for a single solution, which could be easily adapted to a range of different games. / O projeto propõe duas abordagens de geração de conteúdo para missões e mapas em um jogo com calabouços, e um protótipo de jogo desenvolvido como prova de conceito da factibilidade do algoritmo e também como plataforma de testes para experimentos com usuários. A primeira abordagem define missões ao gerar mapas de calabouço em conjunto com chaves e portas trancadas através do uso de um algoritmo evolutivo. Já a segunda abordagem irá popular o espaço do calabouço criado ao preencher o conteúdo de suas salas usando algoritmos de curvas de preenchimento de espaço e autômatos celulares. O algoritmo evolutivo evolui uma estrutura em árvore que codifica informações sobre o calabouço. O objetivo é convergir os calabouços gerados para se aproximar ao máximo da configuração de entrada fornecida por um designer de jogos. O calabouço possui informação sobre as salas, como a quantidade das mesmas, as conexões entre elas e suas posições em um mapa 2D. Também contam com informações semânticas relevantes para a geração de narrativa no calabouço, que são o posicionamento de pares únicos de chaves e fechaduras através do calabouço. Portanto, uma maneira factível para o término do mesmo pode ser gerada proceduralmente. O conteúdo das salas é gerado usando curvas de preenchiment de espaço como as de Hilbert e Moore, além do autômato celular do Jogo da Vida de Conway. Resultados computacionais demonstram que o algoritmo evolutivo é capaz de criar calabouços com propriedades muito próximas às desejadas para uma grande variedade de entradas para calabouços com até 100 salas. Resultados preliminares de validação com humanos mostraram não haver diferença estatística entre os níveis gerados proceduralmente daqueles feitos por humanos. Testes posteriores de validação centrados em humanos com o protótipo de jogo mostram que os níveis gerados pelo algoritmo são percebidos como feitos por humanos de maneira semelhante ou até em maior grau do que suas contrapartidas geradas de fato por humanos, assim como são percebidos como mais divertidos e difíceis do que estas. Portanto, esta pesquisa objetiva gerar elementos de jogabilidade combinando diferentes algoritmos em uma única solução, que pode ser facilmente adaptada para uma variedade de jogos diferentes.

Denna studie undersöker olika designval inom datorspel för att skapa intressanta och varierade nivåer som kan tilltala både erfarna spelare och nybörjare inom en av de populäraste fritidsaktiviterna i världen. Att designa en nivå i ett spel är en tidskrävande process, och ett sätt att korta ner på detta arbete är att låta en dator skapa nivåer efter specifika instruktioner, genom en process som kallas för procedurell generering. Målet med denna studie är att ta reda på vilka designval som behöver tas i åtanke för att skapa program för att procedurellt generera nivåer i spel som är anpassade efter en specifik målgrupp. Studiens forskningsfråga lyder: Vilka element krävs i ett 2D-plattformsspel för att spelare skall tycka om spelet? Samt vilka undersökta PCG-algoritmer och parameterinställningar lämpar sig för att skapa ett spel som uppfyller dessa önskemål? För att göra detta utvecklades ett datorspel som använde sig av olika former av procedurell generering. Respondenter från olika målgrupper utformade efter deras spelvanor intervjuades för att ta reda på vilka aspekter personer från varje grupp letade efter i ett datorspel, och det egenutvecklade spelet utformades efter respondenternas svar. Efter detta så fick ett större antal informanter testa spelet och svara på enkätfrågor som sedan sammanställdes och analyserades för att ta reda på vilka aspekter av spelets genererade nivåer som informanterna från varje målgrupp tyckte att datorn hade lyckats med. En del av de resultat vi fått från studien var bland annat att folk som spelar ofta blir mer motiverade av ett poängsystem än folk som spelar mer sällan, samt att alla målgrupper tyckte om en stor variation mellan nivåerna. / This study examines different design choices within computer games to create interesting and varied levels that can appeal to both seasoned gamers and newcomers in one of the most popular hobbies in the world. Designing a level is a time-consuming process, and a way to shorten this work is to let a computer create the levels based on specific instructions, through a process called procedural generation. The goal of this study is to find out what design choices that needs to be taken into account in order to create programs to procedurally generate levels in games that are tailored for a specific target audience. The research question of this study is: Which elements are required in a 2D-platforming game in order for players to enjoy it? And which of the examined PCG-algorithms and parameter settings are suitable for creating a game that fulfils these requirements? In order to do this, a computer game was developed which used various forms of procedural generation. Respondents from target audiences based on their gaming habits were interviewed in order to find out what aspects people from each group looked for in a video game, and the in-house developed computer game was designed based on the respondents’ answers. After this a larger number of people tested the game and answered a survey, which was later compiled and analysed to find out which aspects of the games generated levels that the players from each target audience felt that the computer had been successful in. A part of the results that we found in this study are among other thing that people who play games more often are more motivated by a score system than other players who play less. And that all target audiences prefer varied level design.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-07-12T11:59:57Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Luca Bezerra Dias - Dissertação de Mestrado (versão final) - Capas Oficiais e Normas ABNT [Digital].pdf: 5112956 bytes, checksum: 8c627192221d904c973b582471a171dc (MD5) / Made available in DSpace on 2016-07-12T11:59:57Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) Luca Bezerra Dias - Dissertação de Mestrado (versão final) - Capas Oficiais e Normas ABNT [Digital].pdf: 5112956 bytes, checksum: 8c627192221d904c973b582471a171dc (MD5) Previous issue date: 2015-08-31 / FACEPE / Os avanços na área de computação musical ao longo dos anos trouxeram no-vas possibilidades e também novas perguntas. A área de geração automática de con-teúdo musical ganhou bastante interesse (BILES, 1994); (SAMPAIO, TEDESCO e RAMALHO, 2005); (GIFFORD, 2013), e com ela, surgiram questões mais profundas sobre o conteúdo gerado por estes sistemas. Qual a aplicabilidade de tais sistemas? Como trabalhar com conceitos como melodia e harmonia? É possível gerar música que não seja apenas um combinado aleatório de notas ou excertos musicais? Reduzindo o escopo da discussão para sistemas geradores de composições percussivas, um elemento essencial fica em evidência: o ritmo. Qual o impacto de um ritmo destoante dentro de uma música? É possível gerar ritmos neutros, que se en-caixem em qualquer composição? Como gerar automaticamente um bom ritmo? Para responder a essas questões, é necessário antes chegar à resposta de uma pergunta mais fundamental: como saber se um ritmo A combina com um ritmo B? A literatura (inclusive musical) sobre esse assunto é escassa e, na sua maioria, subjetiva, pessoal e sem bases empíricas. Em geral, o caminho tomado pelos traba-lhos costuma se basear no que pode ser feito algoritmicamente, para depois encaixar os resultados dentro do conceito de ritmo, tornando os resultados questionáveis. Diante disso, resolvemos seguir o caminho inverso, partindo de teorias e con-ceitos para só depois chegarmos ao algoritmo. Para tal, pedimos que músicos e estu-diosos apresentassem definições de ritmo e suas características fundamentais, a par-tir das quais construímos hipóteses para responder à pergunta deste trabalho. Isto demandou buscar junto aos músicos possíveis hipóteses, encontrar uma forma de representá-las computacionalmente, levantar exemplos concretos (dados) musicais e decidir como representa-los computacionalmente, planejar experimentos que pudes-sem confrontar hipóteses e dados, para, enfim, tirar conclusões com respeito à per-gunta de pesquisa. Neste trabalho, que se insere na linha do uso de computadores como ferramenta de auxílio ao musicólogo, narramos este processo de investigação, elencando as dificuldades, justificando as escolhas, apresentando os resultados obti-dos e discutindo as lições aprendidas. / The advances in the computer music field throughout the years have brought new possibilities, and also new questions. The automatic generation of music content field has aroused a lot of interest (BILES, 1994); (SAMPAIO, TEDESCO e RAMALHO, 2005); (GIFFORD, 2013), and, with it, many deeper questions about the content gen-erated by these systems were brought into attention. What is the point of such sys-tems? How some concepts such as melody and harmony can be dealt with? Is it pos-sible to generate music that is not just a random sequence of notes or musical ex-cerpts? Narrowing the scope to percussive compositions generative systems, an essen-tial element stands out: the rhythm. What is the impact of a dissonant rhythm within a song? Is it possible to generate neutral rhythms, which could fit into any composition? How to generate a good rhythm? To answer these questions, first of all it is necessary to find the answer to a more fundamental question: how to tell if rhythm A matches rhythm B? The academic literature (including the musical one) on this subject is scarce, and mostly subjective, relying on personal opinions and without empirical basis. In general, the path chosen on the studies is based on what can be done algorithmically, to only then fit the results within the concept of rhythm, making the results unreliable. Therefore, we have decided to take the opposite way, starting from theories and concepts, to only then think about the algorithm. In order to do so, we have asked musicians and researchers to provide us with definitions for rhythm and its most fun-damental characteristics, from which we have built the hypotheses to answer this re-search’s main question. To achieve this, we had to ask musicians for possible hypoth-esis, find a way of representing them computationally, find concrete musical examples (data) and decide how to represent this data computationally, plan experiments that could compare hypothesis against data to, finally, get to conclusions about this re-search’s main question. In this work, which is highly linked with using computers as an auxiliary tool for the musicologist, we describe this investigation process, revealing the difficulties, justifying the choices made, presenting the results obtained and discussing the lessons learned.