Here at RISE, we are dedicated in using emerging technologies to drive innovation across multiple industrial sectors while creating social and economic benefit. So it was with special excitement that we welcomed Dr. Panayiotis Charalambous RISE’s Virtual-EUPNEA: Living, Breathing Virtual Worlds Multidisciplinary Research Group leader, to the #ALL_RISE blog!

 

Dr Charalambous holds a PhD in Computer Graphics and Animation from the Department of Computer Science at the University of Cyprus; his thesis tackled the problems of simulating virtual crowd and analyzing their behavior using example data from videos. He later was a Post-Doctoral Fellow at Inria Rennes-Bretagne Atlantique in France where he worked on simulation and animation methods for large scale Virtual Crowds. Before joining RISE as a team leader of the V-EUPNEA team, Dr Charalambous worked as visiting lecturer at the Department of Computer Science, University of Cyprus and also worked as Associate Research Scientist and Computational Scientist at the Cyprus Institute in the CaSToRC research center.
 
With the world currently undergoing an unprecedented event: the COVID-19 pandemic have you wondered how the gaming industry was affected? Leading online games services and platforms are seeing a surge of activity and the latest estimates indicate that the global games market will generate revenues of $159.3 billion in 2020 from 2.7 billion gamers around the world keeping the video game business larger than both the movie and music industries combined! Panayiotis shared his point of view on the matter and provided us with examples of how Gaming helped the revolution of Artificial Intelligence!
 

Welcome Dr. Panayiotis Charalambous!

 

 
Firstly, the increase of activity on online games and gaming in general during this time is very natural to me. Games provide you with a space where you can be immersed in a completely different world,  they allow you to explore and access places where you cannot in real life and offer you a way to handle the situation by entering a virtual world.
 
As for the mentioned notion, this is my opinion as a non-gaming professional expert, rather as a person who grew up playing video games that were story lined based (for the old timers, Sierra On-Line and Lucasarts/Lucasfilm games of the late 80s/early 90s). I am not a fan of games that “force” players to continuously pay in order to basically progress because most of the times they don’t provide you with a concrete story line and they tend to view the user as a money making machine. That is my opinion of course and I understand how many people like them. However, the main reason I decided to study computer science were games that most people nowadays would consider ancient as in order to progress in the game you had to type in moves such as “pick up this object” or “move to the left” or “look at taxi driver”. I never felt that I was used by these games, I felt I was respected, I felt that the worlds I was immersed in and the characters I met were real, even if fantastical in nature. I still remember these games and I still play them when I find the time. In general, I’m not sure if by creating games whose sole purpose is to make money is attracting the “right” kind of people into the gaming industry and I am glad to see that still a lot of indy games – and several AAA titles - are story based and do not fear to take chances in tackling complex matters.
 

 
Technology changed in more than one way. Some companies created game engines to create their games; i.e., a game engine provides the mechanics and basic building blocks to create a game. These companies then realized that it is probably best to release these engines to the world and make them available to anyone as they saw a lot of potential in doing so. It is not only the people that study computer science or people who attend ivy league universities that can create games, its anyone with determination, good ideas and the willingness to work.
 
Prior to that, in order to create a game you had to pay a lot of money so this release of game engines to the world allows people from all walks of life to create games. I find this very exciting and I’m glad to see that over the years companies realized that they should provide people with more “freedom” and equip them with high quality tools to create games and I believe that this will allow for the creation of game types and experiences that nobody could even think of before.
 
Democratizing game engines and resources, making high quality assets available to the world that helped in creating many triple A quality games (meaning games that cost 100 million dollars to create) is a revolution!
 
One very exciting aspect of Game engines is that they are also being used to improve artificial intelligence techniques. People nowadays use them to generate simulations of the real world to train machine learning systems, such as autonomous cars or robots, to behave well in the real world. What was referred to as “kids’ stuff” when I was growing up is now the driving force for achieving what we want in the “adult” world such as autonomous cars!
 

 
What is interesting is that a few decades ago, when people started creating 3D games, some companies realized the need for creating specialized hardware that would allow games to run faster by doing repetitive math operations such as matrix multiplications that are at the core of Computer Graphics algorithms; these hardware are now called  Graphics Processing Units (GPU). What is truly astonishing again, and is quite related to the Games Engines I mentioned before is that these GPUs are now being used to help artificial intelligence more forward at a much faster pace than what could be achieved by regular CPUs (Central Processing Units).
 
Over the past 20 years these GPU devices started being programmable by programmers, meaning they allowed programmers to write software on them. During the beginning of the 2010s, people realized that the math implemented on these GPUs could be used to accelerate what is called Deep Learning, meaning learning out of large amounts of data. Hardware initially developed to accelerate games (“kids’ stuff”) ended up being one of the main reasons we see this surge in artificial intelligence today.
 
If we did not have the gamers of that period and the gamer demands of that time I don’t think we would have the success in Artificial intelligence we have today, I believe this would happen over a longer period of time.
 
I find it very interesting, how two completely different worlds met at one point and decided that this combination was going to create something that would change the world, Artificial Intelligence.
 
Also, in a more Computer Games related aspect, I find it very interesting how the gaming industry brings together people from very different and diverse backgrounds to work on a common vision. When creating a game, you need people with background in music, people that have a sense of storytelling, you need engineers, experts in artificial intelligence, and you also need artists! Such collaborations happen for the purpose of creating a new game!
 
 

 
My research group’s work has a focus on simulating and analyzing life in virtual environments, that includes individual people behavior as well as collective behavior, meaning the interaction of multiple characters such as animals, cars and humans in the same environment.
 
By recording footage of how people behave in the street for example and analyzing that data through algorithms we can train autonomous agents in virtual environments using machine learning and other techniques, to behave like those people. In other words, we are able to analyze spatiotemporal data - data in space and time - in order to create a model of how people would behave in given environments. Once you have that information and you create a good behavior model, you can use it to simulate virtual people in a bigger more complex environment – imagine learning a behavior model from a street and transferring it into a simulation of a city like Nicosia with thousands of people. Why we do? We can use these simulations to create more realistic Virtual Worlds for Computer Games, Movies but we can also use them to train autonomous cars or other robots for example to act in the presence of humans.
 
 

 
As an MRG leader I hope to have people on my team that will go out in the world and do better than me. Researchers that could change the world!
 
As a team, we are currently working on some very exciting projects; some in collaboration with other teams here at RISE, some with companies and some with other academic institutions all over the world. I strongly believe that as a team we will create technology and algorithms that will be used in real world applications, such as entertainment or urban development. Stay tuned for updates!
 

____________________________________________________

 

Don’t forget to bookmark this blog, email to a friend/colleague or sign up to get instantly notified on every new blog post. Of course, we’d love getting your feedback — and what you want to see discussed (and showcased) in the future.