 Hello everybody. I'm here to present Sios product. I'm Xavi Hernandez. I come from Barcelona, Spain, and I work at Sios, that is a company that makes virtual reality for phobia treatments. And we do it with Blender. So that's why I wanted to show to the Blender community how we do the environments, how we do the product, and to explain a little bit what is our history as a startup when we started three years ago. And I will be talking a little about fears and phobias, because there is a lot of people who doesn't understand really what is a phobia. I mean, I didn't know what is a phobia before working with Sios. The vision that founders of Sios had to how to make phobia treatments which we are available for everybody, and talking about virtual reality, of course, Blender. And later on in the presentation, I will do some tricks I learned these years with Blender, how to improve the workflow to develop fast and cheap. It's a similar as a game production. So I will start talking about phobias. I found out this guy on Google, so this guy is supposed to have a phobia. Phobias are not just fears. People tend to think that, yeah, I have a phobia because I'm afraid of spiders. But phobia is not just a fear. Fear is like you are scared a little bit, like me. I'm scared of it, about public speaking. It's not easy. I'm uncomfortable right now, but I can do it. I don't have a phobia. A phobia guy will be here, super shy. He won't be able to talk at all. She will be paralyzed. His brain will be nuts. I met people that had really tough problems on his life here that crashed at his car, her car, because she saw a spider on the windshield. She lost control of the car, and she crashed. So that, for me, is a phobia. When you cannot control your body reactions, when you are facing your specific phobia, there is also a lot of phobias. There is people who have phobia to olives, for example, that they can eat an olive, and they can die because of the nut. So this is a common problem. Most of the people have some kind of phobia and some kind of degree, and these phobias are treated with systematic desensitization. Basically, it means that if you are afraid of flying, a psychologist brings you to the airport and teaches you how to train your feelings or how your anxiety works and explains you how to avoid the negative thoughts. But it has to face your fear in the reality. For example, you're afraid of flying, you have to take a plane, and that is expensive. For example, if you are afraid of driving, you cannot go driving because the tap is kind of hard to control, et cetera. So usually, therapists nowadays use imagination, pictures or videos to make the patient to imagine that they are in the situation that they are afraid of. They also use relaxation techniques. You can breathe in, breathe out, and muscular relaxation, so all kind of tricks that therapists use to teach people how to avoid their fear. But most of the people don't go to treat their phobia because it's not a barrier on their life mostly. When it's really a problem for their life, they go through it, but often it's difficult to see the results because you are not going to face your phobia because you're not satisfied with the results, usually. And virtual reality is supposed to change this paradigm. Virtual reality, there is a lot of studies that proves that virtual reality is effective for treating phobias. But the reality is that only universities does studies. But at the end, they prove that virtual reality is helpful. But this virtual reality doesn't reach the market at all. And nobody uses the virtual reality to treat real patients. They do studies. Most of the universities, they prove it works. But at the end, it's so expensive to bring virtual reality to the common practice that virtual reality was related to only studies. Till 2016, that is the virtual reality boom right now. There is a lot of devices on the market. So there is Oculus, probably you all know Oculus, HTC Vive. Google is also investing a lot of money on doing daydream, cardboard, on your smartphone. Your smartphone can be a virtual reality device right now. And there is a coming, a lot of more, more, more hardware that will make virtual reality a reality. So the vision of Sios was to treat anxiety disorders and phobias easily at home. One of the founders, Danny, on the right, he had a phobia of to fly. He was not able to fly. His girlfriend wanted to go paraglations, and he was not able to go to Tenerife or to Ibiza because he didn't want to take a plane. He was doing boat trips, stuff like this. And one day he said, OK, it would be nice that I could treat my phobia with virtual reality in my home in the sofa, like some kind of tutorial to learn that this can be cured. And his friend, Xavi, they both did the business plan. They are the business guys who had the idea. But they are not developers. They didn't know how to do it. They didn't exist anything on that time. So we are talking about three years ago. So they had this idea, and they put an online advertisement searching for a developer. Then me and Victor appear. So we are the tech guys. Victor is the coder. He does all the back end, front end. And I'm the 3D developer. So I do all the modeling, also graphic design stuff. And the last member of the initial team was Edu, that this is actually his LinkedIn profile picture. He's the karate guy, because he really likes karate. But he's also a psychologist, and he helped us to understand how the product has to be done for a proper therapeutic use. So when I joined their team, one week later, we were selected to go on a health startup incubator. So we are considered health because we are doing virtual reality with a smartphone. So it's an app. So we were selected to go Baltimore. Baltimore is supposed to be so cool, but we are from Barcelona, going to USA. We will work together for a lot of months, and we will learn a lot. But when I reached Baltimore, I remember that the wire series that was like some crazy city, that was we were living in some crazy neighborhood. So it was a really fun time there working a lot. And the gym bullies always say that I was the gym bullies there, because if you guys saw the wire, probably you will understand why. And there in Baltimore, we did our MVP, the minimum viable product. So besides the software, we had the idea to make the software. We were using Blender. We were using Unity as a game engine. And we need the hardware. So because most of the therapist has a normal smartphone, we needed at least a minimum powerful smartphone to run the VR. And that time, cardboard even didn't exist. So we 3D printed our first goggles. And the first kit was a Chinese smartphone, really cheap. It was around 150 euros. The 3D printed headset, that was around 10 euros cost. And with that, we developed the platform. The platform is a website where as a therapist you can control what is happening inside the virtual reality. When you're wearing the goggles, the therapist cannot press the buttons to do interaction. And the virtual reality is very limited. You cannot interact at all because there is no harm recognition, there is no controller, et cetera. So we wanted to give the control to the therapist. So we built a platform to launch all the events that the therapist will address to properly make the patient face their fears. So it's more complex than that because there is a lot of studies behind. There's protocols, different ways to approach the problem. But I'm going to talk about more of the product, not the therapy behind. So there we built over three first scenarios. These are field flying scenarios that is a plane. And at that time, I was alone. I was not able to build characters at all. So we think that there was little heads. We used photographies. So we built everything based on photography. And we did a really low polygonal mode to model this to be able to run the virtual reality on a smartphone. We also did the fear of bugs with augmented reality, with a card you can track, and you see through the smartphone spiders or cockroaches. That was the most common insects that people was afraid of. And we did the first acrophobia. So we did the first acrophobia that was New York Empire State Building that was on top of it. And the therapist told us, I cannot use this. This is too extreme. I should start from the floor and go higher. So we think, oh, let's give them the most hardcore we can. This way, they will be able to cure people. But they say, no, no, this is not helpful for us. So right now, this is one of the highest level. And people don't reach this level because you won't be ever on that situation mostly, especially if you are afraid of heights. And with that product, we reached our first early adopter that was a different manhunt. Thank you. And on Baltimore, we got our first funding that was $50,000. And with that, we came back to Barcelona. After four months, we had our MVP. We learned about businesses, how to do. And we got our first customer there in USA. We had a plan. We needed a plan. We have low money. We are starting to run the business. We cannot spend the money without knowing what to do, and let's make a plan what we need, what we want. We want more funding because with 50K, we won't survive at all. So we need to find more money until the product is be able to maintain the company. So to get more funding, we learned that investors usually want more customers, more numbers to prove that the business will become true and will become profitable. So to get more customers, we need to develop more product to make the better product, make a better platform, more scenarios. Because if you go for a therapist and you tell, yeah, we have fear of flying, we have acrophobia, and fear of spiders. Yeah, but I have very few patients with those phobias. I need phobias with these kind of phobias, these kind of phobias. So we did a survey. We learned about what are the most common phobias. And we make a plan. We needed a bigger team. So we started searching for psychologists, modellers, unit developers, and sales team. Because sales is one of the most important parts of the company. And then we found our first colleagues. They were recent graduates, really young. And they give their best talent they could. And they joined the company. So we grow up a little bit. And we had limited time to build the product. We had to understand what is the difference between mobile VR than desktop VR. On that time, mobile was really limited. You cannot move on the scenario at all. You cannot sit down, look around. You cannot interact. You cannot make doors open. You cannot use your hands at all. So knowing that, we have to make the product so effective we can to don't make that the lack of the mobile VR is very noticeable. We need to make somehow the product shine the environments that you don't need to use interaction. So we have to make static environments. You cannot work on the environments. Because if you work without wanting, you can get dizzy, stuff like this. So I don't know if you guys used virtual reality before. But virtual reality tends to get a lot of business, especially if you are moving without knowing where you're going. If you don't have control, you cannot decide where you are moving. And then patients were getting dizzy. So we decided to make the environments really static. Also, we didn't have enough team to make all the characters animated. So we choose to make all the characters really quiet and non-interactive for the first version of the product. No interaction at all. The user cannot interact. Who has the control of the platform? Who has the control of the virtual reality? Is the therapist on the platform? And we have to understand that the smartphone is very limited. We cannot do a field of public speaking with a lot of people and 2,500 people sitting and moving around because smartphone cannot do it. A computer can, but a smartphone cannot. We decided to invest time on the smartphone because computer hardware was really expensive. And most of the therapists won't afford that for due treatments. And smartphones are quite cheaper. So the basic roadmap was to build the simple treatments to bore the most common phobia treatments and later on start building the complex one. So one year after coming back Spain, we had around 100 early developers, early adopters, sorry. And we upgraded our kit. We stopped doing the 3D printed version. We bought a Omidot headset that was one of the best headsets on that time. And we upgraded a little bit the smartphone, the VQ got an HD screen, so we upgraded a little bit without changing the price of the kit. But most of the early adopters was not paying anything. We was giving for free the kit for them to use and give us feedback on how to learn what you need, what you need to understand completely how they use the virtual reality because we did it in some way because we are gamers, we are developers, but they are not. They are usually very grown up people, so they don't use technology. And most of them was the first time using technology. So we need to make it super easy for them. Without the new team, we develop, we grow up the scenes. This is a public fear of flying scenarios. There is four scenarios for the treatment itself. So there's the waiting at home when the therapist can make the anticipation of the fear when you're going to take a plane. For example, the basic scene contains a crash on the TV. There is a news that the plane crashed. Some kind of stuff that makes people worry about the plane that they will have to take on the future. Then going to the airport. Again, it's anticipation, preparation. So you're going to the airport. You're going to take the plane. And the taxi driver is asking you stuff like, hey, where are you going? Do you know that airplanes, when it's raining, it's more proper to have an accident, stuff like this? And then the boarding gate that you can see for the therapist can choose if it's raining, it's day, night, et cetera. And if there is delays, stuff like this. And the airplane. The airplane itself, basically what the therapist can do is to make air bombs, turbulence, choose the weather. There is storms, not storm day and night. People talking on your site about how scary is the takeoff, how is the landing, stuff like this. We did two more scenarios for relaxation, because usually most of the therapist trains with relaxation. The usual stuff is to build the island. This is not really rocket science. A tropical island is the most common VR relaxation environment that is outside right now. And underwater. Underwater, there's a lot of people who say, common underwater is not relaxation at all, because if I'm underwater, I will feel that I'm drowning or something. But there is a lot of people who likes it. We also did fear of driving to drive on the city and to drive on the highway. So this was one of the most challenging VRs we did on that time, because the city was procedurally generated on the smartphone. And we had really tough moments when we wanted to do the reflections on the mirrors for the user to see what is behind, what if there is other cars coming or not. And later on that time, we built the first VR with animation and with characters. So I will explain a little bit how we did the characters. This was fear of needles that has been used with two hospitals. And they did treatments to people who was not able to take a blood exam. And after doing the treatment with the VR, they was able. So we proved that is effective, that people was able to do it. So this is the first generation of sized characters before we were using, basically, mixamo with destination that libraries. We wanted to do it super real. So we did a photometry based. I don't know if you guys know what is photometry. But basically, photometry is there's a lot of cameras here around me, and a guy press a button, and they take a lot of pictures of me. And with photometry, they can guess what is the volume of myself. But the trick is that the good studios, they have like 90 cameras that they do the picture at the same time. And the tractors needed also facial expression. Going back to the photometer stuff, we didn't have enough money to buy 90 DLRs camera. So to play this, well, this is our workflow. So for example, this is Danny, one of the founders. We needed a casual guy with a backpack. So on the studio, we set up a proper lighting. He was staying here. And I was with the camera doing a lot of pictures, trying to make a lot, a lot, a lot of pictures so much again. If you saw before, there was the cameras. So with the software, we can determine the volumetric. So this is the raw what we get from the photometry. So I guess it's not that really this is open source. You can do it online. You can send the pictures to the desk at 120D, and you get back this model. So the quality is quite good. If the guy doesn't move, sometimes if the guy moves, then there is not nice quality. For example, we have to do two pieces, the body one piece, and then the face in another piece because to get the maximum quality possible. OK, you see. Then this is a high poly count with Blender. Use the sculpt mode. And we clean a little bit. We go forward because it's a little bit longer. Yeah, I need more minutes. So we did our first rig, was not really a complex rig. We did at the beginning, we was doing animation with Blender. And these are more or less the results we got. There is facial expression we needed to do. And we built a basic head with all the shape keys. And then with that basic head, we make another shape key with different shapes of people. To reuse all the work and not do the shape keys for each head. So for example, if you see there, there is a shape key H-A-B-O that is influenced one. With the jaw open, it works. And then we did a few experiments with facial tracking. OK, this is real time. So the idea was that therapists could do talk inside the VR. But at the end, we decided to undo it because therapists was not going to buy the proper camera because we were using deep camera to do this. But this was great for making faster the product. And this was the first version of the platform. We put a lot of effort to make it simpler. Later, I will do a demo if somebody wants to see it. And we'll be doing it real time. So one year after, we went again to USA on San Francisco on a VR startup incubator. So more focused on VR. And we was out of the cave. We spent one year working really, really hard. And when we was there, we were like, wow, there's a lot of people here. We found an investor who think that our product is good and invest on us $1 million on that time. And we came back again and spent to make it bigger, to make the product real. So we hired more people. This is actually the team of Sios right now. We are based on Barcelona, as I said before. And on that time, one year ago, more or less, one and a half years, Gear VR came out on the market and we was able to afford to give for free the therapist Gear VR. So we switched from a crappy smartphone to a better smartphone. We was happy because we were able to add more features. We improved the animation. We started using Rigify and combination Rigify with motion capture systems. We started implementing motion capture on overbreed virtual reality. This is a little bit the workflow. I will do it faster because I'm running out of time. So this is a little bit a scene. So we could add more characters, more animations with better memory consumption. So this is claustrophobia. It's a elevator. You don't see elevator here. So when the therapist can stop the elevator suddenly, all the characters react and they are supposed to make the guy be more afraid. This is another demo that we use facial expressions and we also created a lip sync generation. So the tool is available on Spanish and English. So the guy now has facial expressions. We have a liar that it's talking and the lip sync is automatically generated. So we don't need to make it by hand because there is a lot of dialogues right now. So this is a straight recording from the headset. So this is claustrophobia with MRI scanners. You start on the waiting room and then the therapist can choose when you enter. And you see again this guy. This is one of our developers. So what we did to Sophia Tricks, I learned. I don't know if you're working on game industry, but it's better to use light maps to not bake the lights on the texture. This was one of the atlas we had at the beginning. This is the atlas for flying at home. All the light is baked. And you have to switch to night. We have to have another. This is daylight. This is night light. So it's the same atlas, but with different lighting. But at the end, we used to create the channels. You can have more resolution on the color and less resolution on the light. And you can reuse the color to make the lighting change easily with a set of light map textures. Everything is done by texture. There is no real-time lighting on a smartphone. So this is how it looks without lighting. And this is how it looks with lighting. And I will talk a little bit fast about how we do the animation. For example, I don't know if you guys understand what a little bit you're needing. This is on the left. You see the layers of the characters. This is one of the characters. It has a lot of animations. And this is social phobia. It has to be a treatment that it's 45 minutes. And we have to make 45 minutes animations for five different characters with a really limited amount of memory. So we did like a puzzle. Lot of layers. All the characters share the same animations. And they share the same facial animations also. But they have different lip sync. So for example, one character says something, and another just says something. Then on the head layer, we choose which character looks at what character. So it's really important that the characters look at each other. If you share the animations, they always look forward. But with programming, you can make the character look left or right. And this is one of the, well, for example, you see on the left, the animators did the animation. And later, they animate the scene by itself launching the state of the animation itself. So each character has five layers. And each of the layers says what are they doing? What are their facial expressions? Or what are they looking? And compress everything. As an artist, I don't like compression because it lowers the quality overall. But at the end, you need the product to run smoothly. So the animation is compressed. So it loses a little bit its quality. But you should compress always. And to CPU optimization, you need to reduce polycount a lot. So we use, Blender is really efficient in that way. For example, for the body, you have a rigid body. You don't need to redo the body and redo the rigging. You can just de-semit the body and just break the head. And you can remove loops manually with the head. And you don't lose the shape keys within Blender. That always don't give you that. For example, this is the face. You don't lose the shape keys. You only remove loops. And if you don't need the tractor to talk at all or to have a facial expression, you just use the estimation that it works pretty good. And also, for animation, it's very important to reduce the amount of bones. For example, if I want to make 100 tractors, I don't need 100 bones. If the tractor is sitting down, it doesn't have to walk. So more or less, the future of VR is here. If you want, I will talk deeply later because I'm really running out of time. And I would like to ask Blender Foundation to add some features on the software. I would like a panel that it's created, a button that creates awesome realistic scene. And when you press it on the scene, then it bam, makes a scene that will be awesome for us. So if you can work on these guys, it will be amazing. Yeah, it's done right. Then we have another two buttons that create a realistic character and realistic animation also. So thank you, guys. This is the team. And I hope you like the presentation. So if anybody wants to try the VR, I can run a demo here. I will be here around. So if anybody wants to try it, come to me and ask me. Yeah, sorry, sorry.