The Digital End-to-End Value Chain

Alright, we got everybody on board? Can you guys hear me? Yes? Last 2 days, I'm sure this is the last thing for you, everybody. I know it's been a long 2 days, but hopefully very productive, a lot of things to learn, a lot of things to cover. So as Josh introduced everybody, so it's Renate, Ellery, Joost, Ferdinand, and Lena, and myself, Jazmyn, and we're here to talk to you about a couple of things. Uh, we'll start with introduction about each person here, about what they do, what their role is. There is a, a bit of, uh, talking about what they do at their company as well. But overall, this conversation's really gonna be more about the, kind of the big picture workflow. I know you've seen all these vendors here. They have a lot of amazing tools outside. Um, this is really not about the tools. It's about a workflow. It's about showcasing to you how things can work together. Um, but these elements all can be swapped out. We're trying to show you, um, end-to-end value chain and how companies work together both externally and inside of the brands as well. We think of this as the bird's eye view of how to actually utilize a lot of these tech tools that you see. So without further ado, maybe we start with Lena, introducing yourself.

Hi, my name is Lena and I'm from Browzwear. We'll talk a little bit about the company later on. I think I told our marketing folks the one thing I don't want to do here is to talk about Browzwear, But Yazan has it that we need to do an intro, so that will come.

Great. So my name is Ferdinand. I'm from Visian. And yeah, as Lena said, we are probably covering the very end to the customer side, but also as an example and part of the process rather than figuring out how we as a company are operating.

Hi, my name is Joost from Pixelpool. I think I could share some insights on how We use 3D assets in merchandising and sales, so maybe just before where Ferdinand is operating.

Cool.

I'm Ellery Connell from Foundry. I'm a creative specialist focusing a lot on workflow and tool implementation within existing pipelines.

My name is Renate Eder. I'm from Visu, responsible for business development. We are part of the preparation phase, so to say, so to create libraries focusing on materials, like digital materials.

All right.

Just, I guess, have a seat because it's going to be a long one. So the best way to show this approach is we really actually want to showcase how at a brand level, how does this actually work. And for the most part, it starts with something that looks like this. This design is actually made by Didi Malkaash, who's in the front here. Raise your hand. She's ex-Adidas and now works at Foundry. The idea is that this is what you typically start with from a design brief, right? You get some inspirational things, whether it's materials or sketches, the sketch of the actual product, still living mostly in 2D in this world, and some color variations and some notes. The idea is how do we take this to what we are now looking at, a digital version of this that might go into an e-commerce platform without even selling or making an actual physical product? Obviously, that's our long-term vision. As we get there, you do make different samples at different— timelines to actually see if that product works for you, right? And what we're trying to really showcase is, you know, where do we all fit into this workflow? And we've broken it up to a number of different elements. So acquisition, this is where you actually extract data and we'll go through what different partners and tools that are available today. And if we're missing anybody, let us know. We're gonna be doing this again in Milan. We'll add your platform as a vendor. Again, this is more about explaining how things work together. So acquisition, illustration, virtualization, visualization, e-commerce, and obviously as an asset management system. And that, that type of, um, structure also flows into PLM systems for manufacturing side, as well as supply chain with materials and trim and other things. So again, the idea is to showcase how this would all work. Um, we're gonna go through the actual step-by-step process of what are the platforms, what's available, Again, this is not about what the tool is. It's about showcasing to you guys that there's plenty of different tools out there. It's get started. Choose something, put it in the workflow, and then if it doesn't work out, swap it out with something better. And you learn and you move on. And you're gonna see this type of statement made a lot during this process. It's about failing quickly. We talk about this a lot, but we truly do mean it. It's way better than not doing anything, which we're still seeing This is our set— my 7th or 8th PI. I still see some brands that are— they're thinking about stuff, but really start doing something. Whether it's 1 person or a team of 5 or a team of 10, it doesn't matter. You have to learn from actually using some of the tools that are available that are just amazing to use if you just get started. So acquisition, what does that mean? That means if I have— I start with wanting to get a material. So I saw that design that came through from the designer. I want to now put in some material collections for myself, whether that's a material I develop or a material I source, it really doesn't matter. It's about starting that process. So this could start at the supplier side or at the brand side of things. You've seen outside of here, there's the Visu scanner, which is showcased right here. There are other companies that will use a camera, a flatbed scanner, and like an X-Rite i7. It doesn't matter. Each one of them will give you a different representation of that material. You're looking for the best representation for you. As you all know, how many people have seen a designer that gave you a sketch that you said, "There's no way this thing can actually be made exactly like that." Am I right? Right? You've seen this before. But it looks good to them. It might look good to marketing. It looks good enough. And that's okay. It doesn't have to be perfect. Perfection is not something you're gonna achieve at this point. So once you put it into something like this, let's say the Vizu scanner, it takes that That material scans it, applies lights from various directions on it, and then extracts these what we call maps. And those are the ones at the right-hand side. And those maps basically represent that material's colors, reflectivity, roughness or glossiness, and then alpha transparency basically, and maybe the texture or feel of it, at least in the 3D sense. So we take that, move on to the next in line. So in this case, you'll see some of the acquisition tools available. Again, you're still acquiring, but this time it's physical parameter, it's color, so there's various tools that you can use outside to actually extract that data and save it into your system. And again, it depends on how far you want to go. And once you got those maps and the physical parameters and maybe it's the color data, there's certain file formats that are closed and there's file formats that are open. Obviously, we've heard at past PIs everybody's, you know, We need a standard. We need a standard. There are standards that have gotten started by different vendors. Obviously, on the stage here, we've got Vizu and Browseware that started the U3M. If you, as a brand, are super interested in having a standard and really serious about this, ask for other vendors to support it. Otherwise— and this is no fault of a vendor— if they say a client doesn't ask for it, I'm not going to spend resources on it. So if you want an open standard, there is one right now. It holds physical parameters as well as the scanning data. If you are serious about a standard, ask your other vendors to provide that support as well. We can only go so far. It's an open system. Anybody can collaborate in it. And it's pretty damn good.

Am I right?

Yeah.

All right.

Moving on to illustration. So illustration can be broken down to various things, things like— Tech pack creation. So mostly, I know Adobe Illustrator has been kind of something that's been used. I know Colorway at The Foundry is also another tool. Techpacker, who are here— I'm not sure if they're around. Raise your hand if you're Techpacker as well. Well, they said they're going to be here. Their booth was outside. For graphics, obviously there's a bunch more— Illustrator, Photoshop, Affinity, Sketchbook, Procreate, a bunch of different tools. Designers kind of gravitate to the tool of choice. Again, it's okay, it's just another data set that needs to go in there. It's really about the flow of that material. And moving on to color, there's different, and again, we don't cover everybody, but some of the things that we've seen used, whether it's an Adobe Illustrator, a.ase file, or if it's from Coloro or Pantone, it doesn't matter, it's a color that has a representation in various, either it's RGB, CMYK, or others,.qtxf files that represent them, so it again depends on how you're using them and for what, whether it's print, Physical display, digital, and moving on from that point. And then you have virtualization. And what that means is really creating a virtual prototype that represents your physical prototype. So imagine a real shirt. Now I want to make it into a physical shirt, or that design. There are tools where you model things. So if you're familiar in CAD programs, this is kind of representing when you model things. It's a lot more helpful to do this for things that are like shoes or trim, but then if you're doing fabric, you're better off simulating these kinds of things, and you can go and use other tools out there within that space, and again, we have a number of them out there. Again, when we talked about this is not about which tool, it's about starting. Get started, use a tool, if it works for your team, and you'll find out that some teams actually can use multiple tools. Again, it's okay, as long as they get the job done well, As long as you set some sort of structure or standard internally that you say, here's my deliverable, that's totally fine. And then visualization. So visualization is when I'm rendering something. I'm showcasing something, an imagery that actually represents what I've designed, right? It looks— something looks photo-real, if you will. And there's various versions of tools out there, whether they are ray tracing, or what we call real-time, or WebGL, and if you want answers on any of those, we'll be more than happy to answer questions about what the differences mean in this case. But the idea again is to find the right platform for you. Try. Literally take the list, start one by one, and if it doesn't work, move on to the next one. That's gonna take you a lot less time than thinking about what's gonna happen there. These tools move fast. All the tools that you saw here move quite quickly, and you wanna make sure that you just start now so that you don't end up, oh, it's a year from now, Oh, you've added like 50 new features that I know nothing about and now I have to learn everything again. [FOREIGN LANGUAGE] And then for visualization, you can also export it out to things like custom services platforms that are available today. So not just something that you would do in-house, something that you go and look at retail visualization beyond just the product. What do I do after the product itself? And there's more of that as well. And that also flows into commerce. And it's a bit of a blurry line where you're seeing some of the visualization happen at the commerce side as well. And again, there are companies out there that represent what types of services and platforms and products, uh, that are provided. And again, a lot of them have booths around here. So if you want to talk to anybody, they're here. And if not, I'm sure there's additional ones. I see them in the audience kind of, uh, sitting down. But, and then finally, uh, in this case we're talking about digital asset management. So digital asset management, Unlike PLM, where you're talking about the manufactured product, this is about the digital asset that you're working with, the 3D file, the, the garment, the pattern that's still in a digital form. There are tools out there, whether it's for product management or for material management. Obviously for us with Swatchbook, we do materials, but there are others that do some materials and overlap as well. Again, find something that works with what you have right there. And the last part, which is again, the suppliers and mills, we're now see— actually starting to see adoption of suppliers wanting to be a part of that digital conversation and uploading their data to platforms in material space and engaging with the brands that way before they actually send them the physical samples. And finally, uh, the product, the PLMs, I mean, you're still using this in the production pipeline. A lot of, if not most of these platforms have API, uh, uh, systems that can actually integrate into PLMs directly. So talk, talk to them about that. Talk to 'em about how we can put this into my existing infrastructure or system or move into a, a new infrastructure. That was an overview of kind of all the tools that you're able to do. We're actually now gonna take a deep dive into kind of what we were doing with that design we had in there. But I'd like to open up— this is more of a conversation. We have 67 minutes to fill in. That takes some participation from you guys. So we'd like to open up some questions first around what you saw now, and then maybe we'll move on to the next stage, which talks about the product and how our tools were used in that product creation. Creation. Any questions? I think Joost had a great idea. Was it Joost or Ferdinand? Who was the one that gave us the— Joost. It was basically, what is the number one nagging problem that you came to P.I. thinking, I wish there was a solution for? Let's start there. Look at something like that and ask a question about it. And if we know the— are you already raising your hand? And if we know the answer, we'll give it to you, and if we don't, maybe there's somebody else in the audience that can actually Chime in.

I've noticed a lot of the 3D rendering technology is kind of lacking in the sweater area, specifically in being able to render different stitches, not just one that you already have. Do you see any progress of that coming in the future?

All right. I think the question was about stitches, right? Yes, so in the area of stitches, I think there are companies like Substance. So there are two, maybe there are two ways to answer this. Firstly, there is a way of how do you mimic the look of a stitch, which companies like, I think Substance have come up with really real to life kind of stitches which you can apply onto any of the software and then be able to render it. The other part of it is the actual construction of a stitch, whereby you actually have a pressure point on that. I think that is something that is coming. There are workarounds in each of the 3D modeling tools that allows you to add pressure to the edges so that you are actually able to mimic that behavior as well. On the area of sweaters, that's a good question. I think area sweaters has always been— if you look at it today, a lot of the software for sweaters is actually connected to the machinery. right? And a lot of them requires what we call the knit technician, right? And that's probably something that requires you to have a PhD and a physics to get into that. So I think that that is something that will evolve over time as it's not just the software, but I think what's important is to be able to translate a lot of the bitmaps and then into the gauges and then flowing into the machine. So as part of an end-to-end, I think when you take something that is in concept And you want to move that into the manufacturer, then the machinery guys come into play. So I think I see out there somewhere, Shimaseki is here, Santoni, probably, you know, Stoll. Those are the 3 big guys. And I think at some point it will be up to many of you in the forum here to bring them to the table and say, I want to be looking at something that I can take my designers right through to the point where your machineries can take that far. Is that going to be far? Probably not, right? But that's a conversation that is in the hands of many of you that are here to bring also the machinery guys. But I think from a software perspective, everyone is ready. We are very open and waiting for the connection to happen. But I think it takes both hands to clap. So hopefully that gives you a bit of an answer. Okay, cool.

We'll get Sean.

Hi, Dave Sharma. Great. I commend you for compiling this list. It's very informative. One of the questions— one of your descriptions, which was light in that description about e-commerce. All right, now you've created this 3D asset garment itself. It's got all the fabric properties, got sizes, grading, and all that.

Can you just raise the mic a little bit?

Thank you.

How do you now— push that 3D asset to the consumer itself online? Because online is increasing at the expense of in-store. But how do you challenge the return rates of 30% plus on online? And how do you challenge the conversion rates which are stuck at 2% and 3% online? And so the e-commerce part is where my question is focused.

So I'm going to go to maybe Ferdinand or Joost. Both can start. We'll start with Ferdinand first.

All right. So I mean, you address two major problems, especially when it comes to e-commerce. One is the return rates problem. The second one is the conversion rate problem. So return rates is like you essentially need to figure out how your customer is looking like and do it on an individual basis rather than on a statistical basis. So there are solutions out there that can measure the body very accurately. And reassemble a 3D model of that very specific person. And then, like, this is also what we're going to talk about, like, today is, like, how can we reuse this asset and really create a virtual dressing room for online stores. And this is essentially what we are doing. So, like, we are using assets that have been created in the development process and push them forward to the customer. And re-simulate it on the human body itself. So first of all, they don't only get the size recommendation, they can have a fit recommendation based on their own personal preference. So they see how the garment is fitting on their body. And that ultimately also is to a certain extent solving the conversion rate problem because you take away the fear from the people that basically want to want to buy certain items because you take away the fear of like, is this sort of item going to fit me in terms of size, but also in terms of style. So like you can cover both, both sort of ends if you really pull through the whole supply chain here.

I think another interesting aspect there is that if the product ends up at the e-commerce website, that means it went through for some brands at least, it went through a whole cycle already of assortment planning and design, the design process. And if there's a large return rate, that might mean that the product doesn't really fit the current market. There's maybe not just enough interest for it. So apart from making sure that the products actually fit the people that want to buy them, I think there's a huge role for 3D in speeding up the whole process so that you can correct react a lot quicker to what the market wants so that whatever products you do produce fit the market a lot better. And you can also increase your conversion rate by having more relevant products.

Can I make a comment?

Of course. Yeah, so Lena had a comment.

So I actually have a quick one also to add to what Jost is saying, that when you look at returns and so forth, right, we can look at it from a front-end part of it about e-com and selling, you know, how much can we sell, how much can we return. But at the end of the day, what's important is to close the loop, which is importantly, how do you take those information on what has been returned, why has it been returned, and how do you feed that back into your design, your development team? I think that's where the loop really closes. You cannot get convergence, you cannot get that returns down on a cycle basis. It has to go over a period of time. And one thing that we're not doing right now is we're not closing the loop. It was very interesting that we were, we were, my, our CEO and also my sister was in the talk. And one of the things that Amazon did in the last 2 years was to create this foundation, I'm sorry, this platform for global business selling. Now what do they do? They go to a factory, literally, and they say that, hey, you know what, I'll let you create a brand, I'll let you sell it on my platform, and what would I teach you? I would teach you from statistics that I have gathered, right? I will tell you things like what has been returned. So the case study was a puffy jacket for a woman. I would tell you in a puffy jacket, long jacket for a woman, what are the 3 main things that has been returned, right? Not enough pockets, zipper qualities are not good, sorry Tucker, and then, you know, really also the bottom hem is too tight. Now based on these 3 things that Amazon is giving to one of these factories, they corrected, created a jacket that last year in 6 months on Amazon sold 3 million pieces. Was there a return? Likely not. But this is where I think is challenging our industry, not just to look at e-com and your e-com team as a standalone, but to bring that information on return sizing and fit, because that's what's really going to drive your product creation in order to reduce in the mid-long term all the returns and also the sizing issue.

And actually, I Another comment on top of that is, I wonder how much of that actually gets relayed back to the designer. So it might go to the factories and the development, but shouldn't the designer also be informed like, hey dude, that design wasn't that great. Maybe you need to change your approach or change the material, what you've selected. I think it's a learning process that, as you said, has been disconnected so far. And data is greatly useful if you know what to do with it and you look at the patterns and you extract something from it, but it needs to be looked at by specific people in the organization in the process, and they have to have visibility in that. And designers look at things differently, development looks at things differently, manufacturing looks at things differently. Just because you give them, here's an Excel spreadsheet, doesn't mean they can make any use of it if they don't know what they're looking for. So that's another part that I actually think— that's a really good comment there.

Sean?

You're right behind my thing, I couldn't see you.

Just a quick question.

It really doesn't like you.

So just a quick question. You showed us a whole bunch of tools up there.

A bunch of them.

These tools oftentimes when I'm taking a 3D I'm moving it from one to another and to another, I lose all sorts of components of that. Of course. What types of things are being done?

That's a feature, not a— that's not a bug.

It's a feature.

Right.

I know there's competitive landscape, but what are some of the things that are being done to kind of help that be a little bit more seamless, especially with things like Unreal Engine and Unity? And, you know, I want to push a button, send it right up there to be able to take into an AR environment or whatever it might be.

Maybe actually it's a good opportunity to start going to the actual presentation to see the next steps are. So a part of this is to also showcase how some of that— nothing is perfect. It's not always seamless. You're absolutely right. You still lose some data, especially when— how do you deal with stitching from one application to another? It's different approaches. So that definitely needs to be addressed more. I think we're going to— let me go through the process a little bit. And I think then we can go back to your question. And I would love for each of us to kind of ask and answer what are the things that we see from our side? Because when we get data, what are the challenges for us? When we push data, what's the challenges to push that data out? Make sense? All right, but great question, which leads really good into our presentation. So we're starting off with Vizzu. So Renata, we'll be able to talk about what, you know, who is Vizzu, what they do, and a little bit more about the process for specifically the product that we were building.

OK, so Martin and Reinhard, the two founders of Visu, they were both a part of—

Can you raise your voice?

I should raise my voice?

Just raise your voice. Not after two beers. I know, everybody's got a different voice.

So Martin and Reinhard, the two founders of Visu, they were both part of Adidas, actually, German sports brand, and part of the 3D virtualization team there. And they were especially responsible to research on how to get digital materials. And at that time, they found out that there is not really a solution or technology out there which can do that at a large scale. So basically in 2013, they decided to leave their job and to found Visu to pursue this, or to fill this niche or this gap which was there in the 3D process. And I think we now see that they really kind of met the industry need, because we can say that the materials scanning technology we have is definitely a leading scanning solution. Thank you for going back one slide.

Want to go back?

Yeah.

Sorry.

Thank you.

Because towards the third topic here is that we currently sell to not only the fashion brands, footwear, apparel, but also to automotive. And we see also a growing demand from the architecture and interior furniture industry, and our devices Devices are in more than 25 countries now. Now you can take it.

Now it's good? All right, sorry.

So basically, the idea is that the digital materials feed the full 3D pipeline. And ideally, you do the digitization as early as possible so that you achieve a constant quality along your digital product creation pipeline. Development, marketing, configurators, e-com, just to name a few. Should I talk to it?

Yes, talk through your presentation.

So this is showing a tiny piece of our core process, so to say. So you have a stack of materials. You have an existing material library. So it's not about material design, but to digitize existing materials, basically put it into our hardware, and I think most of you guys have seen it today or yesterday already. Then device is taking images and deriving texture information, surface information from it apart from, yeah, getting color, transparency, transparency, roughness, and so on. We have tiny features improving the process, meaning that if you have not ironed your fabric, we kind of do a certain wrinkle removal, which looks funny, but you need that more often than you think. But basically, it's to digitize, to get your existing fabric digitized, to get it kind of flat smooth and then to get a seamless repeat. You can do this seamless repeat of course in Photoshop or any other tools, but you can also use our software using kind of more intelligent functionality, which is basically analyzing here on the left-hand side the structure and then getting a seamless output for it by preserving the physical size of the, of the fabric. And this is basically the— this is it. As easy as this is.

Cool.

So that— so speaking of that question, so one of the things that we worked with VIZU where they grabbed our API and they were able to take that and directly from within the VIZU software you can actually go to file.

This is your part now.

Yes. And go directly to Swatchbook. So in this case you actually can upload the data to Swatchbook without actually leaving the environment that you're in. And obviously this could be the same person, it could be Different people across the globe, it doesn't really matter. You basically put some initial tagging and then you take that to SwatchBook. We, from that point on, move on to our side. So we're founded 2016. Our materials, basically the idea is it's a single place to manage your materials and the materials coming in from supply chain as well. We obviously, as a brand, you might be dealing with an entire supply chain of materials, but you might be developing your own. You want one singular very— what we're looking for is a visual, intuitive system that allows you to, on mobile, on iPads, or your browser, to quickly go see your materials, make your decisions on them, put them in collections, and then get out. We started our Swatchbook service center in China just last January. We've been scanning about 1,000 materials per month. So we have about 5,000 and growing. And the idea— all these materials have come from from existing brands on our platform today, which means that they've been vetted with these, with these clients. Obviously, every brand will have to vet them for themselves, but at least it's a starting point to get everybody to use materials in 3D or otherwise. And to give you an overview of what that looks like today, this is what Swatchbook looks like. So whether you're on a browser or a tablet or a phone, we showcase materials in a visual manner. This is called one-to-one. It showcases the actual size of the fabric on your screen. You see all the details in 3D, in 2D, in the browser or in the app, even simple videos showcasing how you interact with materials, which is so super helpful to design, right, and sourcing. It's a simple thing. Another aspect is our Capture application, which allows you to quickly take a picture of something, put some simple tags. These tags actually associate to 3D assets, and then you can see those assets directly in your phone, which is in your pocket at all times. Time. So if you see something inspiring you, or you're at a material show, and you want to see what that looks like on a product, you can. It works in 3D as well as in AR. And that's kind of like a 45-second overview of what Swatchbook is. There's a lot more to it, but that's about it. So we take that material that came from Vizu. In this case, we might not have the exact puffer jacket that was provided by the designer, but we do have a bunch of generic data there that we can utilize. And in this case, I actually want I want to grab an ASE, Adobe ASE file with Pantones in there, and I want to create, render and create variations of this material knowing that I like that material, but I want it in a different color. Now, not only do I see that material inside a swatch book, I can see all the different color variations on a product that looks similar, but the idea is that this is step one. This is when I select my materials, right? I visualize them, I like them, I then can move on from that point to actually create my design and apply the same materials that have been scanned in the lovely VisuScanner and see them and visualize them in 3D. The fact that you're able to do this on a browser or on your phone is a part of what makes this special, right? It makes it— I don't have to go in and log in to something that's on a desktop at work, and that's the only time I get access to my materials. We create a collection of these materials. Again, you— once you create that collection, collection, you're able— we'll suggest that that collection is something that you like. You'll be able to quickly add them to that collection as you hover over them. Once you're done, you can actually see that collection, and you can download the data based on your file preferences individually. So I might be somebody who just uses Illustrator for tech packs. I might be somebody who just designs and wants to do a couple sketches in Photoshop, or I could be a 3D designer who uses something like Browzwear, and I want to download download all the materials. Once you do, we have a sync application. You download those materials into your system. Now I have all that data, all the scans that came in from Visu in various colors, and I could move it on and across it to my next partner.

Yeah, okay. So this is, this is Browzwear. So compared to my, all my younger colleagues, Browzwear is like the oldest kid on the block, but at the same time we are—

It's experience, experience.

Yeah, we We are the youngest kid on the block at the same time. If anything else, being the oldest kid on the block, I'm not sure that we really know everything on what to do, but we certainly know what not to do. So the first thing that we always tell people is that Browzwear was not founded to do 3D simulation. We were founded to do virtual fitting room because we believed in '99 that everybody was going to buy on the internet, and when they do buy on the internet, everything will be returned to day Dave's point earlier on this 30%, 40% will be returned and nobody will be confident. So BrowseWear set up, my two partners Avihai and Norm set up to create a virtual fitting room in 1999 where today your mobile phone is more computer, more powerful than the computer we had to import in to do simulation. So long story short, we were very excited. We went to a small company called Benetton and we gave them this whole idea. That small company looked at it and say, oh great, this is 'This looks good, but how are you going to get the garment in?' And we said, 'What do you mean? There are so many 3D tools out there. There is 3D, there's Maya, there's— everybody is out there.' And the guy looked at us and goes, 'I make clothes, not car. Until you figure out how to do it, don't come back.' So this is really what set Browzwear on our journey to do soft cloth simulation. And we've come far from that 20 years having been taught by companies like Delta Galil, having been taught by Adidas, having been taught by Nike, on everything that we needed to know because they saw the simulation being much more important for the creation phase and not so much for selling at that point because they said that we are doing 144,000 SKUs. We make a lot more samples. So why don't you guys use your technology to good use for the world rather than try and sell? So that's really where— and fast forward, Today, Browzwear is a company that's also a new kid on the block in a sense that we have a complete new suite of product working with a very different vision. A vision of seeing that the industry go from a 'I design, I make, and I hope to sell' to a place where we need to design, sell before we make. And this is the vision that has been driving us the last 7 years. This is drive our roadmap. This drives our ecosystem of partnerships that we have today. My colleagues here are the ones— questions from Sean about how do you interconnect, how do you integrate, those are what we live and breathe for today, right now. So we don't have a one-tool solution, and we don't believe that tool is going to help you make you successful. We believe that the tool is only going to aid you if you figure out what that workflow is. And I think some speaker from full beauty brands and so forth, talks predominantly also about workflow and the importance of that. So fast forward, where do we want to go, right? From 3D vision to reality. There will come to your point of time where, yes, I understand everyone here talks about, you know, we need to save on samples, we need to get faster. But I assure you that what keeps many of your CEOs awake at night is how do we sell more? Because if we don't sell, there is no sample to make. Make, right? So this is where we are, um, and that's Browzwear. So going through— yeah, just click, just click this. Uh, yep. Okay. And this is why we believe that we're not going to be doing this alone, right? This is our open platform slide. Browzwear created an open platform about 4 years ago in order to invite other partnerships to come into a world to change fashion industry. We're not going to do this by ourselves. My colleagues here are just a fraction of the other people that we have. There are many out here in this room that have also integration with us. And through various ways from the value chain, from the upstream to the downstream, you are going to be using these assets again and again and again. And BrowseWay is here to just say that continue to take that asset. We'll do our best to integrate with everybody. And we'll keep the formats open so that everyone can You can take that asset and take it downstream or move it upstream. So you see that we have plenty of partnership from people who are doing voice of customers. How do I take my digital garment that has been created by a designer? Let's take it to the market. Let's hear what the market is saying. They love it. They want to do something more with it. And let them communicate. Let them tell you as the brands on how to interact with it. Then take it downstream on, you know, now can we ray trace it? Can we bring it to social media? Sell it, and so forth. And then take it down to merchandising. How do we assort with it? That beautiful jacket that has been created by Didi, you know, what other outfit that needs to come along? There are many pieces of different technology out there, and I hope PI Apparel has given you a view out there on the many people that are coming into our industry. And so Browzwear believes that this is, this is where we need to go, and we could— we are committed to keeping that open, listening to every one of you talk about the importance of openness and standardization, and that's what we are committed to doing. And so when we look at a workflow perspective, we had the designer create something really beautiful. The designer does not need to start from scratch. They take something that they already know, a block. They are using a simple pen tool, much like Adobe is working on. Um, go ahead, slice it, and then from there make the changes. Very intuitively, very easily. Let me go ahead and click on a zipper, all right, a smart zipper, and then move on to all the attributes and the elements of working within that block, right? So again, this designer is not starting from scratch. They are actually creating from something and moving it on. And she's extremely creative, so she's done something magnificent out of this very simple jacket. She draws a line, clicks simulation, and it becomes a puffy line. So we're not going ahead to create many things. Here again, sleeves design doesn't need to be straight. Once that design has been finished, right, we can now go ahead and see it in various mode— hood up, hood down— and just to understand what are the different various color combination that we can work with. And then sending it out to other ray tracing tools that are inbuilt within in the tool to see what it looks like. So this is browseware. OK.

Awesome.

Thank you. So moving on from there, obviously, we still have some additional tools that might be taken over. So I'm going to bring up— so Ellery, you can start talking about your process and who's the Foundry, what do they provide, and move on from there.

Yeah, cool.

And how, obviously, this ties into our system. Yeah. Let me know when you want to move to the next slide. Si.

So Foundry has a background and a history in visual effects. [Speaker:CHRIS_WILSON] Which is why the first slide, or the first picture, is things blowing up. We don't intend to blow up your products, but that's where we come from. But because of that background and history that we have in visual effects, as we've moved into the design areas, we noticed that the software application Modo, the 3D application that we have, was gaining a lot of traction in design just kind of organically. It's a good tool for modeling, for rendering, a lot of different things. Starting to gain a lot of traction in there. And as we started approaching the design area, we're seeing that one of the nice things is that in a lot of design in apparel and footwear and that kind of thing, there is this new move towards 3D. And there's all the discussion of how things are going to be put together. How do we manage the workflow? And the nice thing is that isn't something that hasn't been done before. In visual effects, in media and entertainment, that kind of thing has happened a lot. Now, that doesn't mean that every solution that was created for a feature film is going to be the right solution for creating apparel to sell, and to visualize, and to design. But it means that the underlying idea of how to connect and put the technologies and the pieces together in order to get where you need to go is not something entirely new. And so we've leveraged a lot of that background and that history in order to help design good solutions to help bring in the other pieces, the other partners here on the stage, and put them all together into working packages to take existing workflows that are within— already existing within design and helping to build in the new implementation of 3D visualization, of virtual and augmented reality, and all those kind of things.

Next one.

So this is actually a shot inside of Modo. So you would— in this case, we're taking the jacket that was already simulated and created, bring that model into Modo, where we can do different kind of contextualized rendering. So you can really get an idea as to how a design is going to look, not just as it sits plain by itself, but really in the context of where is it going to be worn. This is the kind of thing where you could even get as fine-tuned as to say, this is a jersey design that's always going to be worn by soccer teams. It's always going to be seen under stadium lights. How do we see how that material and how that article of clothing itself is going to react in that environment? And then we can make intelligent design decisions based off of what we learn in these really early visualizations. So this allows us then to have really good detail and understand things down to how much is the fur on the interior of the hood going to affect the way that the light is absorbed into it? We get some really good understanding of our objects before we actually even even create them, and that helps the designers have a good visualization early on. It helps them to internally fine-tune those design decisions. That helps them to pass those on for internal messaging, for sales, and for anything else. The next one. So then another part of the equation here that we got into is Colorway. And so this is a quick walkthrough video of Colorway that shows the ability to take those renders that are being done And instead of just being working with specifically with real-time rendering or things where you lose some fidelity, these are now based off of real— we call them offline renderings, renderings that don't take milliseconds to calculate, but take a fair amount of time. But because we're using that as a base, we can then go in and make more design decisions on top of that. You can do all sorts of things like see the orientation of your fabric and rotate and move things around and find out how things are are going to best fit to make the pattern work well for your designs. A lot of other kind of interesting things that also happen inside of there, if you look at the very bottom of the screen, there is a little spreadsheet. And the spreadsheet in this case is really just linking up the colors and the materials to the pieces that I've got on the jacket. But that can be hooked into a PLM. That could be hooked into all sorts of data so that you get real-time feedback to the data that is flowing into the material choices that your designers make. That means that they can see basically any kind of information that you might have in your PLM flowing into the design. And so they're making that— again, they're making design decisions with more information than they would have had otherwise. You can see the location of the vendors for the material. You can see relative costing, all sorts of things that maybe you would not have immediately at your fingertips. And the nice thing is, is that it takes one thing out of the hands of the designers, and that's something that I know all designers love doing, and that's making spreadsheets. And it does that for them.

Oh, somebody got that, okay, good, I'm glad I got a laugh out of that.

Thank you. But it will take and automate that process, so there's no more of your designers having to make the spreadsheets, and then there's no more saying, "Okay, I created 8 variations of this, and I forgot to do the spreadsheet on the 6th one," and then we don't have the data, and you gotta go back and forth, refine it. So by taking those pieces out of the hands of the designers and allowing those to happen automatically, it helps them, again, improve and increase their ability to message their design forward. It helps them to refine it internally, and it helps them to bring that out to other applications as well.

Awesome.

I think that was— oh, yeah. And there's another shot of the jacket.

And again, a part of that— more of that collaboration. So we've built a bridge that goes into It takes it to colorway directly, so you minimize spreadsheet editing as much as possible. So moving on to— we have Pixelpool. So, Joost, you can go.

So Pixelpool originated about 20 years ago as an architectural visualization company. And about 15 years ago, we started moving more into the retail industry. So we started visualizing almost exclusively for retail, so we were visualizing stores, and after some years, we also started visualizing the product that goes inside of those stores. So year over year, that part of the visualization process became more and more important to us. So we were working for, well, retail design teams, store design teams, merchandising teams, and eventually also sales teams. So back in the day, those would still be still renders or in some cases animations.

That's all about the point. That's all about the presentation, by the way. We talked about VFX, this is audio effects.

Exactly, exactly. So, well, we were making still images and in some cases animations, but about 5 years ago, we thought, well, the real-time rendering industry, especially Unreal Engine, it got far enough where we thought, well, this is starting to match the visual quality that we want to offer our clients. So we started to move into the interactive space. And we had all these 3D assets. We had all these 3D stores, 3D products, fixtures, all of it. So we could recreate an entire store in 3D virtually. And then we started just thinking about what can we do with all these assets? What can you do if you have an entire store virtually? What kind of interactions do you need? And that started— well, it started like a whole wave of requests from our clients. We started building interactive showrooms that big brands use to sell their assortments to their buyers. And I think that's something where that really resonated with me as well, is that I really like the idea of designing a product and then immediately selling it. So our experience has been mostly in B2B, so not as much B2C. So we are experienced in creating showrooms that brands can use to sell to their buyers. So yeah, let's play the video. This is D-Tail. It's a platform that we've created to power our 3D applications. It's a web-based tool that allows you to import your product line, to import 3D garments that either you or a vendor or Pixelpool has created for you, and start assigning those garments to your product line. You can also create assortments, so those assortments will flow then into each of our 3D interactive tools. And then what's also cool about this detail platform is that we integrated a tool for people that don't necessarily have access to a very beefy gaming PC or don't have a showroom in their office, but they still want to see the value of 3D and especially They want to see how these 3D assets can help you to visualize your assortments in a store environment.

It's okay.

So here we're seeing this product that we just were shown and how this translates into our process. So you see it in a detailed platform. Here we're configuring the product for our automated merchandising platform. So we're saying this is a product that we want to show up in a certain fixture. It has a certain priority in the assortment. So you can influence how it turns out in the eventual render. So this is all web-based, and it communicates with an Unreal Engine-based cloud machine that you can send a request, and it sends you back a proposal of a merchandising view. So you actually see the product here merchandised in a render. So you don't need any gaming PC for this. It's all happening in the cloud, and I think this will really— increase the adoption. So if you want a bit more control in the merchandising process, we have tools for that as well. And I think there's still a lot to be said to keep the control with the visual merchandisers, to let them play around and use their human cognitive abilities to really create the most compelling setup possible. So by dragging and dropping products and fixtures, you can create these setups that match well together and then present them to together in a larger store environment. So here you see an existing store that has been collaboratively filled by different merchandising and sales teams. And we're adding this last final piece of this garment that was created in this pipeline among the partners. So here you see, I think this is a great way to show how the 3D garment can move all the way from design all the way to selling. In our case, this is more of a example of B2B selling, but it does show how powerful it is if you show it in the context.

Awesome. Thank you. That's pretty cool. Ferdinand, to kind of close it off from Fission.

Oh, great. Yeah. So if you reach that point in the whole process line, you have already achieved quite a lot. What I'm telling you is not the end of it.

One more, one more step.

There's one more step. I think it's also a good opportunity to— there's lots of pieces coming together and lots of people working on different pieces. And it's also quite good if everyone sees what is happening before everything is shown to the customer. So you really have the way to— or the means to show it to the customer. And like what we do here is like at Fission, we are a company, software company founded in 2015. We are based in Switzerland. And what we create is actually two different sorts of technology. One is virtual dressing room experiences that are powered by a mobile body scanning application that we also provide where we sort of can reconstruct the body in various different ways. So a little bit of background, how do we operate and probably where do we bring the personalization in. Obviously one part is mobile body scanning applications, but there are also other ways to create 3D profiles and we are partially working with other vendors together, we are also partially providing other solutions, so like there's a combination of things going on on our side. The important thing is like once it's going consumer-facing, the consumer is in the middle. And it's the central piece around which we build our applications. So obviously, once you have the body measurements, once you have garment measurements, you can do all sorts of things. One of them is giving size recommendation. We actually also showed with Browser together a case where we made made-to-measure garments. That was just a couple of weeks ago in Frankfurt. But I'm not going to go into much details about that. It's more about the virtual dressing room today. So we created 3D assets, we want to reuse it at the customer end. And probably a little bit just before we're going to dive in on the virtual dressing room, I think there are also other applications that probably also some vendors here or some companies here are focusing on. So think about personalization, or once you do personalization at the very end of the customer. Don't only think about clothing, it can also be other fields that can be targeted within your company, just like talking about fitness, health, security, and others. Can we go on? Let me quickly introduce like how mobile body scanning applications can be done. So we do it rather simple, we need 2 pictures of a person, like one being from the front. And like you can also do this alone from home, so you don't need to do it together with your partner or with a friend. So you can also do it alone. The features of this sort of body scan is really that we can very, very accurately describe every human being. And describe in physical matters, not in psychological matters. Psychological matters, so that's the thing that we don't cover. What we do here is then create a 3D body avatar and reuse this in virtual dressing experience. And not only on the virtual dressing room consumer-facing, we can also feed that avatar back to the development process. And for instance, if you have a boutique tailor that wants to have customized clothing for a specific person, they can also take this scan and and then really construct the whole thing on the body itself. So partially, just following on in the discussion, this has been a process where the whole 3D asset has been generated in Browserware. What we do is we create a plugin in order to export it and then reuse it on our platform. So what we can do here is we can change different models, we can change different colors, we can also change sizes, and we actually also give size recommendation based on the body avatar that is in there. So really personalized experience on the consumer end of things. Yeah, I think like here it's important to say like there's like a lot of things that need to happen in the beginning, but I think like we showed a great deal of things that that are happening beforehand. And also, there are different ways. There's one way we showed today, but there are different ways how those sort of things can be generated. I think at the end of the day, every brand wants to sell something. And we are opening up the 3D channel also to the consumer and really make it personal. Thank you very much.

That's kind of the overall. So thank you for everybody. That was great. I know a lot to digest. Maybe we'll go back to Sean's question and just have them comment a little bit about the integration aspect and how do you move data. Does that make sense? So I think we had touched up a little bit about that, so I'm going to carry the conversation. There are certain things that we can collaborate on that moves data smoothly, and some things are a bit more difficult. As a software provider or a platform provider, you're always thinking about the user experience from our perspective. There's different ways to think about what that looks like, and moving data back and forth, again, depends on what is the value you bring to that platform or that software. And it's about, is the value in the data? Is the value in what you're doing with the data? So it's always a tricky question, I think, for every vendor. It's not easy to say, oh, I'm going to give you— yeah, you make something here, spend 5 minutes here, and then do something else somewhere else. I mean, obviously, we'd love to have you in our platform, or as long as possible with as many licenses as possible. But that's— in the real world today, it's really becoming very clear that if you looked at, at least in my opinion, if you looked at the software, the entertainment world, if you— movies and games, they have a role which this industry does not have, which we call the technical director. I know John there knows those people pretty well from Foundry. Their job is to build the glue that moves that stuff across from one platform to the other. It's almost like a natural, it doesn't, they don't think about like, oh, you need to move the data across. They're like, no, no, there's somebody that does that. That's their job. I think that's something the industry has not had particularly that do that type of work. We had this discussion before. It's not ideal. You do have to bring somebody who maybe not even design, doesn't have a design knowledge for this, but that movement, it does really matter. It makes it for a much better process. But what does end up happening is that as a brand, just like a movie studio or TV studio, you specify a standard for your process. And from that, you say, OK, I don't care what tool you use. Use the best tool you can as a designer or an artist. Make your product, but give me ABC. As long as you give me that, you can use whatever you want. I'll help you build some small plugins and export publishing and things like that. But you take that and you move it on to the next thing down the road. Whenever we see at least some opportunity of like, hey, I think we can alleviate a lot of pressure from that movement, there's making that plugin, we can work together. But that's been my experience from my opinion, at least.

From an industry's point of view, I mean, everyone knows this. I think for many of you in PD, you know this other file format that's called DXF, right? 25 years and we still cannot get that DXF correct. From DXF transfer from one to the other. I mean, raise your hands if you tell me you don't face this problem. We don't see any hands. Good. So I feel that— Is that the one where nobody knows, like, do I raise my hand? Yeah, it's like, don't raise your hands. I'm kind of like, you know, because otherwise it will be overwhelming. But I think with 3D assets, it will be very much the same, right? We come from 3D assets from some formats that are different. But now you have very sophisticated technology that are coming up. Great texturing, digital swatches, new formats on storing material. All of these are going to change the pipeline that your work— and it's going to put some pressure, I would say, on Browzwear as well. Like, you know, hey, when your 3D mesh fits into this environment for digital assets, I need you to do this. Now we've got to go figure it out, and, you know, how do we constantly make that change? They could be scanning a body, and we say, you know, you it will be an OBJ. But the truth is, many of you will come and say that, hey, I want this animation, I want that, you know, and so forth. And we would have to work together to figure those things out. I don't think, to manage expectation, that we're going to be there tomorrow. But I do see more and more in the last 3 years that this entire industry are more willing to come together to create that format that takes it from one pipeline to the other, or one platform to the together. It will come, and it's changing day by day. So don't be afraid of asking that question when something from Browser doesn't work, or, you know, you have something for Vision or whoever else is out there and you need it to work, and we'll figure it out and we'll, we'll partner to, to make it work together. But let's manage expectation, and DXF is our example here.

Anybody else wants to comment on that before we move to a different topic?

I would add one little thing, and I think it's important to, uh, to consider the end goal and the solution that you're trying to get might not be the one that you're thinking of. You know, if, if you need data to flow through from, uh, one application to another or one part of the workflow to another, I think it's important to step back and, and think about what it is that you need to get out of it so that you're not architecting solutions for something in a vacuum where you're creating solutions essentially for problems that you may not really have.

Yeah.

But to really sit back and think of what's the end goal that you have before starting to, uh, move forward with more more solutions. That way you can make more intelligent decisions on the solutions that we create.

That was good. All right, moving on to additional questions. I saw some hands raised on the left-hand side. Do we have any microphones? I mean, the ideal thing is you sketch out a little napkin, then you just like take a picture and the product comes in 2 days later, Amazon Prime, right?

Yeah. No.

Thank you very much. Again, Dave Sharma. One is I just wanted to further develop on what Lena said about the feedback loop from, you know, from the consumer itself of an actual garment that goes out into the consumer's hands. One is, of course, you can get a manual loop actually making— getting comments from them. The other is to go back to fission. If you have a body avatar and you have a digital visualization, you can show heat maps, you can show the distance between the actual garment and the body, where it's loose, where it's tight, and so forth. And that creates a great feedback loop for down the chain, back into designing and other things. And second question? I have a question for you, Vishen. I'm sorry, I didn't catch your first name. But that photo-generated avatar, what kind of accuracies are you getting, fidelity between the actual human and the avatar that you generated? And are you actually conducting e-commerce through your site or through your deployments?

Sorry, what was the—

I think the question is how accurate is the actual body scan and is the platform actually used for the e-commerce side of things, am I right? Yeah.

It's a question, sorry, two questions.

Two questions.

Sure, let me answer quickly then. So it's a great question, it's a very valid question because like it all comes down to accuracy. Like so if you want to describe a human body, do it good or don't do it at all. So it's probably one of the most important questions that have been— like that everyone is asking us. We are down at the level of 1 to 2 centimeters. On certain measurements we are better, on certain we are a little bit less good. That being said, we require form-fitting clothes to generate our body avatar. That's probably quite a hurdle to do so, but we don't see that going down without having that requirement. We are working on solutions that sort of don't restrict a person to do so, but we don't expect the accuracy to go through the roof with that. As for the second question, where do we come from? And are we live on consumer-facing applications? Our app is in the App Store. We are just like this virtual dressing room that we just saw is going to be live in, I think, like 2 weeks from now. So it's just in the integration process. And then a couple of tests going on. We did our homework in the— workwear and uniform space. So we gathered a lot of learning there and really trained our system based on the knowledge that we generated there and really put all the puzzle pieces together really to create this experience that we can provide today.

Okay.

All right.

And thanks.

We had another question over there and I think there's another one in the middle. Yeah.

Hi, I'm with Victoria's Secret. Um, so one of our pain points, and then just talking to other intimate companies that are here, are really trying to unlock bras and like fitting bras in the system. Um, I know a couple of companies like Browzwear and Clo have, um, stated that they're still trying to find a way to get a, a soft form avatar. How far out do you think that is?

Did you understand the question?

Yeah. Yeah. So I think there are— the issue is not in the simulation of the garment on a soft body. The question is the availability of a soft body avatar. I think this is something that some people here are trying to crack the code. There are university researchers and so forth that are trying to crack the code on that, and also industry players that have white space innovation that are also participating in some of this. So I anticipate that in the near future, within the next 2 years or less, that some of this, you will hear some of the news that are out there. Yeah.

Yeah. Also probably a little bit more about that. So it's about soft tissue deformation, right? So it needs to be like, okay, how is the body interacting with the, with a garment or like with an object, and that's like currently a hot topic in research, and as Lena said, like, will be like their first results out there. It's just like also very expensive on the computation side. That means like that the loading times are just like through the roof at the moment, but expect that to be quite nicely done in the future.

Yeah, that's an interesting question because It's funny because that also touches on materials, which people bring up but don't think about. Once you start getting sock body deformation in there, that could push— basically, is the material shaping the body, or is the body shaping the material, or is it both? But you also see things like material transparency increase as you— like for leggings and things like that, that might actually get affected. So even the scanning needs to also catch up with that type of stuff so we can actually showcase what does that material look like on a swatch versus when it's worn. I mean, the same kind of things are happening in leather with pull-up and things of that nature where leather changes the actual shading on it once you pull it. Those types of things I think are very interesting and I think are the next step of where a lot of this technology is going to.

And another word to that is we have seen companies do the implementation in this area space, whether it's tight garment compressions and so forth. They have a workaround on that. I'll be happy to share a little bit on that. That's interesting.

There's a question right there.

Yes, hi. I wanted to ask Frederic a question about the virtual dressing room. I'm working with a think tank out of the University of Maryland on a see now, buy now application for the runway, because I have an event production company, and so I want the user in the audience to be able to buy something Something Off the Runway with the app. And we're trying to interface it with that virtual dressing room concept because you're buying it and do you like it, but does it like you back? But my question is about the face. I notice you have faceless dummies, kind of that look.

How—

I'm talking about the uncanny valley, you know, about the— resemblance of the avatar to the actual person. Does that really make a difference in how they're shopping and buying? They want to see themselves. And is it kind of creepy when you see yourself as a robot avatar?

Emotional purchases. I mean, that's what it is, right? That's what you're asking?

Right.

Because I'm thinking if I do the form, do I have to have a face on it and scan the face of the customer?

It's a very good question because there are different opinions in the market out there. And even if you do A/B testing, there are mixed results on this very specific topic. So current state is we as a company, we don't really focus on the face. We sort of partner up with companies that focus on the reconstruction of the face. If you ask me my personal opinion, It's going to be— it needs to be a little bit better to look a little bit more human-like. And also, like, the renderings that come with it needs to be like subsurface scattering. They're like technologies that need to be supported that are not like that easy. I think like there's a lot of development in this area. And like we are like— we have two partners in this area where we sort of look into the details on on what we do. The one is more on the abstract way of reconstructing the human body, just like have a probably a little bit more abstract way. The other one is probably more on the detail. Mixed opinions, also in A/B testing, we don't really see defined results going in one or the other direction. So to be very honest, it's an open topic and we sort of chose the most abstract way to, to, to, to, to put a hat on a person.

There's some interesting technology coming out now that if you've ever used Snapchat and you start looking at the AR filters utilizing your camera, I would imagine there's a lot of technology out there that allows you to showcase the virtual garment but then project your face on top of it. If you've seen also there's some, what they're called deepfake videos of projecting people's faces on somebody else's head and there's one, there was a really good one of, I think it was Bill Hader doing like an Arnold Schwarzenegger. If you haven't seen it, it's deeply funny but it's also very disturbing at the same time where it switches the face and you can't, like, what the hell happened to his face? But that's Arnold talking. I think that's something that we're gonna go through because I think that type of stuff can flip over the uncanny valley aspect where you have to build everything in 3D and make it look real real, it's very difficult for the most part. A lot of artistic input in there. With a lot of this AI-driven technology from AR facial capture and deepfake video type abilities, you're able to put that and marry it together. And for the most part, it works well because as you said, purchases are emotional and you're going there to see, I want my mirror, right? I want the product on me. How does it look? Looks good, doesn't look good. I don't, sometimes I don't want just a product, I want an outfit. So those things, I think, this is just the start. We always think of it like this is just the beginning, guys. We have a lot to do. But yeah, that's a super good question.

And I think the reality of it is, I think, how long ago was this? Maybe 5, 6 years ago, Intel did this whole immersive shopping experience. Everything that you're talking about was actually uncovered also in that experience, that people are not ready to see themselves and so on and so on. So I think to be pragmatic about it right now, It's to do all those A/B testing. You know, how do you sell clothes that is emotive, but yet also try not to make it very distracting, right? So if the face is distracting, if people are, you know, are seeing my face on Arnold Schwarzenegger's body and so forth—

It'll become a meme really quickly.

Yeah, it will become— I mean, it would distract and it will stop you from selling. So start from small ways, less distracting, focus on the garment, and then create other things in the brand. Browsing side of things. Yeah.

[Speaker:COLT MCANLIS] There's a principle in animation also that typically says that psychologically when a viewer is seeing an animated character that the character itself should be slightly less realistic than the environment that it's in. And that allows the viewer to more easily ground the character. And I think the same kind of thing then would apply to an avatar, where if you're going with a less realistic avatar but having more realism in the clothing and more realism in the environment that they're doing, which doesn't come into that uncanny valley issue, then it's going to allow the consumer in this case to more easily ground the product that they're purchasing with something that's real.

Exactly. Awesome.

Any additional questions? The nagging issues that we talked about before? Oh, there's— OK. Hi.

So with the virtual dressing room, I noticed that one of the garments looked it was kind of offset. Is there like a styling, like, thing that you can do where you have like customized positions of things? Like if you can pull it off the shoulder, or is it just in its one standard position?

So it's feature request number one.

Feature request number one.

Basically the question is, if the garment is simulated and it seems like it was off, is that intentional? Can a user kind of almost become like an art director and say, I want it a little a little bit off the shoulder and things like that. Again, feature request number 102.

Of course it was on purpose.

It's a feature, not a bug.

No, it's like there are certain things that can be improved and like to be very honest, we had limited time to put the whole demo together and like there are quite a long bucket list on things things that still need to be improved to really make it perfect and seamless. I think it's a process there, and we are at the step where we can move very fast, small steps, and really make it great at the end. And therefore, those sort of things we only figure out once we really work together with various different vendors and various different companies to really see, OK, like, yeah, we need to improve. And that is probably also going to solve another problem. And we take it step by step.

Oh, no, they started 90 after.

Hi.

My question is more for Lena and for Renato on the U3M. So it's really exciting to see new file formats coming out. But how can we guarantee that that is an industry standard? Standards, how has the acceptance rate been and the adoption? So what's the feedback on it? How are other companies viewing it? Yeah, just developing on that.

Thank you for this question, Luisa. Yeah, so I think that the feedback was generally very, very good when we announced that first in PIA Milan, like in October, right? The adoption is, I think, very good. We have very various vendors who have already implemented that. And it's not only amongst 3D CAD for apparel. We have a 3D CAD vendor for footwear who implemented various rendering solutions supporting that, either natively or via a plugin, which than we did in these cases. But we also have other platforms in this ecosystem, like also Swatchbook, the product, and so on, who support that. And the list is growing. So we have constant requests to get that implemented. Having said that, I know that there is still a lot of work to be done. So we have thrown it out there, so to say, and we see what's coming. So I hope it's— I hope it's there as long as FBX. I hope that it's like with the few—

No, we're learning from the FBX. The FBX effort, we're not going to go down that road.

Then that, yeah, we are currently— we have learned a lot from these first implementations. We are working now together with our partners on the next version, on an improved version. And this is again a call to all of you to participate here more, even more. We know that it's not complete. So we have still this issue of not having physical standards, physical parameter standards, and so on. We are pushing for that. We are working on that. And I think we had another— when we had this call for the prep of this meeting, there was this statement that at least you start from somewhere, and you do one step, and you see what's what's the next. So I think we have put that out there. We see that the next step is to get this physical issue done. And I think there are so many requests. I think it was like every second visitor at our booth was requesting for that. So I think we all understand that this has to be done.

Yeah. And as a vendor or as a brand, there's a GitHub repo that you can be a part of those discussions and input what features do you think are important here. Whether it's about the structure. So it is an open platform for everybody to contribute to. The earlier the better. That means there's more voices. So again, from a vendor perspective, please participate. It helps everybody and you. And then from a brand perspective, please inform your vendors that you're— who you want to have that standard be adopted by, to push them and say, hey, I need you to put this in there. And I think that both sides will get there hopefully and make it productive. Happen faster.

Just one small point to add to it. Remember that the U3M was not a great idea, you know, while drinking a beer with Renate and we came up with it, okay? It didn't come that way. The whole PI Apparel, ever since the beginning of PI Apparel, everyone has been talking about material, importance of material, a standard that, you know, you can, you can take that material and you can plonk them into different system. I think this is our first step towards keeping that open, making it happen, and But as Renate and Yazan say, this requires the industry to come and collaborate together. We've heard your voice, we've started something, but the journey is something that we collaborate and do this together. Yeah.

Yeah.

It's, again, it's not about perfection. It's about, it's not yes or no. It's not black and white. You need to go in there and, and just start something. Sometimes you take a step back, learn from a mistake.

Yep.

That, you know, from that beer discussion, um, and you take one step back, but then you have that informed inform your next step, right? That's— we keep talking about this fail faster, but it's really about not just failing and doing something else, learning what not to do now and move on. Whether it's standards, whether it's what tool to use, whether it's how to work on your workflow, who you're hiring, from what department did they come from, those types of things, you're gonna go into bumps. You're not gonna get the perfect scenario every single time. For those that you have— who have done that, that's amazing. It's, you know, you're a unicorn here, but for the most part, there's a lot of things that are not gonna work out. And that's completely OK.

You also discover new things after this first step. So you cannot just complete the task because at step 0, you just don't know what step 2, so to say. So I think it's good that we just started and now we see what else is needed. I think that's true for everything.

So we have 2 minutes left. I can't believe we actually almost finished 90 minutes. But if there's any more additional questions before I hand it off to the final remarks. We can do a little—

Follow-up question about the U3M standard. So I think a lot of us probably work with mills, and we want to know all of these fabric properties before we actually purchase hundreds and hundreds of yards of them, and then do our own Visu scans, and then scan the physical properties as well. So I guess my question is, how much are you working directly with the mills to create the U3M files and then also integrating physical properties in the future as well as part of the standardized format?

[Speaker:DAVID_KOHLER] I can only speak to the ones that we collaborate with, with the mills. Until now, it's been harder to convince the mills regarding physical properties than it is the visual ones. And somebody has to pay for extracting that data and providing it, right? Whether it's the brand or if it's the, the actual suppliers or the mills. What we've seen, at least from our perspective, is that they're more reluctant to go down the physical properties road. But that has to do with the brand. If a brand comes in and says, "I'm not gonna take your material until you give me the physical properties," a supplier will comply. This was the same discussion 3 years ago that we had on the visual side, where we started asking suppliers, "Why wouldn't you want to scan all your materials and have everything in digital?" And their immediate response was, "If a brand doesn't tell me to do it, I ain't doing it." They don't see the same benefits that you guys do for the visualization and digitalization. They're now starting to understand that when I scan something, I don't have to go to a material show right away. I can release my collections earlier on and have a much more meaningful discussion when I am on site and I do have the physical materials because now that my brands have already seen it, they've seen it on a 3D product, they've put it on, on on their simulation properties, but it is a little bit more of the brand saying, "I need this. You need to provide it to me." I mean, the cost is gonna— somebody's gonna pay for it. It's just either hidden cost or upfront cost. That's my perspective. I don't know if anybody else has anything else to add on that.

I think it's not only to the brand asking for it because we see, I think, I guess 80% of our customers definitely sold their supply already to to do that. It's also that I think you as a brand, you have to ingest the data and really use it. Because when we talk to suppliers, it's not that a supplier does not want to do that or doesn't see the benefit. They see a lot that they do this as an additional task, but there is no— like, the data is sent, but it's kind of lost, or it's not productive-culottes.

Well, what's their ROI, right? It's what's the value coming back from it?

From that.

So, yeah.

That for us, if I can speak for my situation, is that when we're working for brands and we're creating these showrooms and we're creating a lot of products for them, I think there's still a lot to be improved of the people that are actually putting out these requests with the vendors. They're not necessarily the people that are working with the mills and working with these standard files. So it's not propagating yet far enough within the organization. So it's not the salespeople and the merchandising people that are aware that it's actually really important to share these properties so that vendors like Pixelpool can assure them that the product looks the way it should look, right? We're working from material that it's not precise enough, and we would really like to have—

Yeah, we all would.

—a brand that just gives us a whole zip file of U3M files and just lets us visualize those. That's perfect. So I think that's a really good idea, and I think it needs to propagate more within the organisation, not just with design. Absolutely.

And I think most of the time, whether it's at the mill or at a factory, the reality of this is, you know, the journey of 3D at the factories have always been a lot faster than the brands and so forth. Why? Because, you know, creating samples, creating heirloom, you know, samples and your bills and bills and bills of fabric has always been just a necessary evil to getting the business. If there were other ways of doing it, they will sign up for it. And if we look at presentation by Kohl's yesterday, right? If we are dumping things, 11,000 pieces every hour, something is wrong and breaking in our workflow, right? So I would not worry whether the mills are going to sign up. I think they would. The question is, in collaboration, what is the trade-off? What are we giving them back to change the entire way the industry works? And that will be the one thing I add to it.

Yeah, it's a good thing to finish up on. So our time is up. Thank you so much for everything. Closing remarks by Josh still.