Inspiring Look into the Future of Artificial Intelligence and Game Engineering

Once again this year, students on the Master’s degree program in Artificial Intelligence Engineering (with the two specializations “AI Technologies” and “Game Engineering”) presented their final projects at the “Showcase Evening: AI and Game Master”.

The number of participants in this event increases steadily every year; this year, around 150 people attended, including around 80 students. Each of the 11 student groups was able to present their projects at a table with a large screen. These included AI applications, self-developed games and AI and game crossover projects.

Fast identification of luggage at the airport

Impressive projects from the AI Engineering Master’s program were presented that evening, including “Deconvolution”, an innovative game based in a world generated entirely by AI. Another project called “OnHands” uses computer vision and machine learning to enable the rapid identification of luggage using smartphone scans in cooperation with Austrian Airlines. The fact that two students are working directly with Austrian Airlines on the baggage identification project shows how closely the students are already collaborating with companies from the business world.

Another impressive VR puzzle game was “That 1 Wizard Game”, which creates an immersive gaming experience through skillful wand control and rune recognition using a neural network.

Winning projects: “Ultravision” and “No Way Back”

“Ultra Vision”, in cooperation with Meduni/AKH, focuses on identifying faulty ultrasound probes using CNN-based approaches, implementing a binary image classification model based on VGG16.

At the end of the evening, the two winning projects of the evening, “Ultravision” (AI category) and “No way back” (Game category), were honored. The guests were also treated to pizza and beer.

Bernhard Knapp (Head of the Master AI program) and Alexander Nimmervoll (Head of the Game specialization Master AI) were delighted with a successful evening together with the students.

“Impressive variety of innovative projects”

“The impressive variety of innovative projects at the ‘Showcase Evening: AI and Game Master 2024’ reflects the outstanding potential of our students on the Master’s degree program in AI Engineering. The evening was not only a platform for creative technology solutions, but also an inspiring look into the future of artificial intelligence and game engineering at UAS Technikum Wien. Many thanks to all the students for their outstanding contributions, which exceeded our expectations,” said Bernhard Knapp, Head of the AI Engineering degree program.

Diverse areas of application in the Master’s degree program in AI Engineering

Diverse and promising areas of application open up for students on the Master’s degree course in AI Engineering. Artificial intelligence is seen as one of the key concepts of digitalization and is expected to influence numerous aspects of our everyday lives. People are already interacting with intelligent bots on the World Wide Web, autonomous vehicles are no longer a rarity and intelligent household appliances are increasingly finding their way into our homes. In the gaming world, players compete against smart computer opponents in automatically created artificial worlds, while doctors are supported by advanced software when analyzing images. Furthermore, texts, such as chatGPT, and images, such as in Stable Diffusion, are increasingly being generated by AI.

On the cover, at the top: Alexander Nimmervoll (Master’s in Game AI, left) and Bernhard Knapp (Master’s in AI, right) with the winning team in the Game category: “No Way Back”.

Picture-Gallery:

Here is an overview of all projects from the current year:

Deconvolution

Team: Markus Goldgruber, Maria Schildbeck
Category: Game
Description: Deconvolution is an immersive game in a completely AI-generated world based on the Gaia hypothesis by Lynn Margulis and James Lovelock, which states that the Earth can be considered as a single large organism. On the journey, players connect with objects and thus become a complex organism themselves. In the end, you arrive on a planet and the organism you have become triggers its unique unfolding and you become a planet yourself.

GetRasty

Team: Anna Matt, Thomas Verweyen
Category: Game
Description: GetRasty is a tower defense and real-time strategy game in which players defend their castle from invading enemies by strategically placing towers along the predefined path. In the game, you manage workers who autonomously perform tasks such as resource gathering and tower building. The game mechanics revolve around placing towers, managing the supply chain and understanding the interactions between different tower and enemy types. There are three different towers in the game, each requiring special ammunition to be produced by workers.

midas

Team: Alexander Pallisch, Momen Abdou, Adnan Vatric
Category: AI
Description: Analysis of the stock market: As part of the Development Project, an AI trading bot was programmed in Python, which is connected to MetaTrader 5 and can trade completely independently on the forex and stock market. This bot integrates comprehensive risk management and follows a moving average (MA) trend strategy. In addition, we have developed a neural network that analyzes financials and makes predictions. This network is an important part of the bot and improves its trading decisions through professional predictions.

No Way Back (Winner Showcase Evening: Game – Master 2024)

Team: Dario Bachinger, Rainer Laschober
Category: Game
Description: In No Way Back you fight as a single character through turn-based battles in arcade style. Choose the way you want to solve each of these battles, there are more than one option. But beware: with every victory, your chosen approach becomes more difficult to implement in the next level. Switch weapons, collect points, acquire upgrades and plan your tactics from far away or up close.

Swordmasters: Samurai vs. Viking

Team: David Staudinger, Gerald Hirsch, Alexenia Weibel
Category: Game
Description: “Swordmasters: Samurai vs. Viking” is an energetic real-time strategy game set in East Asian or Northern European environments. With a focus on fast-paced combat, players can choose between attack and defense and place armies instantly. The 1v1 format allows one player to attack while the other defends, offering both PvP and PvE scenarios against human or AI opponents. The game’s standout feature is its procedural map development, which allows for infinite playable worlds. With visually appealing environments and quick turns, it challenges players’ strategic thinking and creativity.

OnHands

Team: Bushra Yasin, Alexander Kreissl, Philipp Ehm, Michael Schwingshackl, Marvin Elias
Category: AI
Description: More and more people want to travel and discover the world. Our luggage is our most loyal companion on all these trips. Unfortunately, suitcases and bags are increasingly being lost, which not only frustrates travelers, but also puts a strain on Austrian Airlines staff. Entering this data manually can take over a minute per item of luggage. In the OnHands project, we use computer vision and machine learning to identify the type and color of a piece of luggage scanned with a smartphone in a matter of seconds. This information is automatically entered into an international database for lost luggage.

Spellslinger

Team: Adrian Dimaano, Peter Helf, Thomas Neumann, Christian Prior
Category: Game
Description: The SpellSlinger project is an immersive, entertaining and challenging VR puzzle game. Here you play a wizard who has to complete various puzzles and obstacle courses by skillfully wielding a magic wand. The focus of this project is on the playful execution of spells. These spells are recognized as spell symbols in this game by reading the gestures of the wand as runes. This rune recognition is performed with a neural network, in a unique use case of AI in video games.

Ultra Vision (Winner Showcase Evening: AI – Master 2024)

Team: Paul Hassanpour, Manuel Kranzl, Saifur Rahmani, Michael Zeiner
Category: AI
Description: This project aimed to develop an application to identify faulty ultrasound probes based on test images. Focusing on the implementation of a binary image classification model, the tasks included data exploration, file standardization and extraction of relevant image regions using deterministic and CNN-based methods. A model based on VGG16 was selected for classification. Preprocessing and data augmentation were explored but showed limited effectiveness. Hyperparameter tuning guided model optimization. The final application, implemented in Python and converted an executable file, deals with the identification of faulty ultrasound heads.

BelAIcik

Team: Gerald Kloimstein, Peter Szabo, Thomas Ville, Alexander Zotter
Category: AI
Description: BelAIcik is a video analysis software created to help coaches and scouts automatically analyze low-quality footage from a single angle. It includes several AI components for field detection and play recognition and uses classical computer vision approaches for the rest. BelAIcik transforms side views into a bird’s eye view and calculates some statistics along the way. The prototype has shown that some information can also be extracted from a video view.

D-Shifters

Team: Yannick Helling, Dorian Bach, Stefan Held, Philip Stibl
Category: Game
Description: This science fiction game takes place in the future. The player has a group of players at his disposal with whom he visits different dungeon levels containing loot, loot chests, enemy encounters and exploration of the map. Combat is turn-based, but also dynamic (no grid movement).

Dungeon Gourmet

Team: Lukas Hartinger, Nico Winter, Philipp Lakheshar
Category: Game
Description: Dungeon Gourmet is a rogue-like action RPG realized in the Unreal Engine 5.2. In the game, a “meow-ster-like” chef, armed with a giant cleaver and his throwing forks, sets out in search of the ultimate recipe. To his surprise, however, he encounters never-ending hordes of enemies, with each surviving wave granting him one of three bonuses that make him stronger on his further journey.