In this episode, we explore colour forecasting in the metaverse. We first establish what we

mean by the metaverse and colour forecasting before exploring the opportunities

and the potential disruptions that colour forecasting in the metaverse may have

on fashion trends, both in the physical and virtual world.

The Global Challenges Podcast

Transcript; Colour in the metaverse – episode four

[00:00:51] But first, I'd like to introduce our excellent panel of speakers today, Professor Steve Wesland. Would you like to go first?

[00:01:34] Peihua: Hi. I'm Peihua Lai I am a current final year PhD student in the School of Design. My research is about fashion image analysis. I have developed an algorithm to automatically extract colour information from fashion images. So images used in my research include both physical outfits and virtual styles in the metaverse. I believe this research will benefit fashion and forecasting industries.

[00:02:35] Caroline: Thank you all very much for those great introductions, I think we've got a really nice mix of experts here on our panel today. Steve, to start, I wonder if you could just give us a quick introduction to what the Metaverse is, I think the metaverse is probably a word that some people might be familiar with, but others may have no idea what that is.

[00:03:28] Caroline: Yeah, absolutely. I think you are right there. I don't think there'll be one. I think we may see these connected platforms which we may end up calling different Metaverses. I think it's also worth adding to that that, whilst we dpmt have an established or unified metaverse at the moment, we do have these virtual worlds already that have been established.

[00:04:13] So hopefully that brings us nicely onto forecasting, we're talking about fashion and colour forecasting today. So Sally some of our listeners may not know what we mean when we talk about fashion forecasting or colour forecasting, and obviously our focus today is more on colour forecasting.[00:04:31] But would you mind just giving us a bit of an introduction to trend forecasting and in particular colour forecasting?

[00:04:59] You know, [00:05:00] more and more now we talk about our world from a global level as opposed to local. So all of these things have to interconnect and combine. So when we talk about colour forecasting, yes, absolutely, that's what we're talking about today. But you can't talk about colour forecasting without talking about trends, themes, aesthetics, they all go hand in hand, and I think that's what's really important to sort of stress here.

[00:06:07] And so it's a really important point, even when we think about the as well as we're talking about today, to sort of think about how our predictions of colour is going to influence the world around us, including the virtual.

So I wonder if you could perhaps share some of these challenges, or changes that you've seen that companies may have experienced within the fashion industry. I wonder if we can start unpicking those links between forecasting and the manufacturing supply chain for fashion and textiles.

Some companies have dramatically reduced those times, and this has advantages both economically and in terms of sustainability because in an ideal world, we don't actually make anything until it's already been sold, and that would dramatically reduce so-called dead stock and the mountains of textiles going to landfill.

[00:08:50] Caroline: Thank you Steve. Sally, can I just quickly come back to you here? You know Steve's talking about supply chains becoming a lot shorter.[00:09:00] If we talk about traditional forecasting industries, do traditional forecasters normally work to, is it those kind of 18 month long predictions or is there that flexibility to be shorter?

[00:09:35] I definitely think there's a need for that and there's an absolutely an argument that that's going to then reduce the over consumption and the overproduction of products that maybe won't sell. You know, there's a long, long conversation about how you go into a store and you see, you know, a rail of garments in a certain colour that why are they in the sale?

[00:10:24] You know, fashion traditionally would be four collections a year coming out and so trends were in line with that time scale. So I would argue that we're not trying to say, let's forget the past. We're trying to say how can we improve it and which industries can benefit and be improved by using a quicker timeline and which still need to look two years ahead.

[00:11:12] Caroline: We need to remember that we often see, you know, macro trends, micro trends. So there is a place, for these longer and shorter timeframes. Peihua I think this is a great time to come to you as your PhD. Your research, you've been exploring some of these ideas [00:11:31] and in particular, I know you've been looking at using current consumer trends and we might come onto this nowcasting you might want to refer to that, but so you've been using current consumer trends as a means of consumer driven colour forecasting. So rather than it being way ahead, months, as you say, Sally, you know, 18 months, two years in advance, we are looking at a much more immediate way of forecasting trends [00:12:00] and colour. [00:12:01] So perhaps you'd like to start by just telling us a little bit about your research.

[00:12:56] Instead of just the dominant colour [00:13:00] in the images. So therefore the company in different sectors can use more accurate and detailed trend information for their business, such as what colours of hats are most popular and London this week or for this month.

[00:13:34] I think that's really shows the power I suppose, of data and how you can take a data approach to trend forecasting.

[00:14:04] we could argue that the pandemic coronavirus, we didn't have any clue what was going to happen there and how that was going to influence the market. So then that would change what we call nowcasting, which would absolutely have an impact on the trends that we've predicted for 18 months or even 6 months ahead.

[00:14:28] Caroline: it's a conversation, I suppose, isn't it, between all these different things and that trends aren't static. It's not as though we say, yeah, this is going to be the trend next year. As you say something like a global pandemic may come along and completely change the kind of social economic factors around that time and hence the trends.

[00:15:40] The other thing to say about the metaverse, at this point is that the metaverse or virtual worlds, many people think of those as being all about gaming. We're not just seeing people using virtual platforms forgaming, we're seeing it very much as a space for people to [00:16:00] socialize, as a community space, a way for them to express themselves perhaps, you know, through what they're wearing, what they're doing, who they're connecting with.

[00:16:31] And we have a big movement I think at the moment towards Web three in a more democratized fashion industry within that virtual fashion industry. Taking all that into account, do you think that the metaverse and fashion within the metaverse and trend forecasting within the metaverse could change our existing forecasting industry?

But paradoxically, I think that's not the case. My view is that some of the trends that are currently developed may be artificial and that if instead we monitor consumer behaviour to identify trends new colour and fashion trends will still emerge, but perhaps not changing quite as often as in the current. I would say expert driven colour forecasting process. So one of the reasons Caroline, we've been looking so closely at the metaverse [00:18:00] is, as you say, it's really easy there for consumers to exercise choice. I definitely think there's even more argument in the metaverse for consumer driven trends than in the physical world.

I would like to think there will be synergies between what happens in the metaverse and what happens in the physical world and this certainly could accelerate changes towards consumer led forecasting. I do like the term you use Caroline, democratized fashion. So I think the changes that might occur as a result of the metaverse might not only affect trend forecasting, but in fact the whole of the fashion industry.

So Sally, we've heard some interesting perspectives on how virtual worlds could be a vector for change, certainly when it comes to consumer led forecasting. I'd like to now ask you your perspective on the relationship between physical fashion and forecasting and virtual fashion and forecasting. Do you think that the two things may influence each other?

You know, we talk about digital trends that the combination of digital and physical and you know, we are living around all of this information, when we think about color forecasting or trend forecasting, we're, we're taking in so much information that includes things such as AI and developments in machine learning and the metaverse and it's influence on trend forecasting.

You can look at Pantone's Colour of the year for 22 called Barry Perry, and if you read about that, that’s very much influenced by digital world and the metaverse being one of those, and it's this idea of one influencing the other and the back and forth. So we've seen lots of trends coming into the physical world, even in accessories and stationary, for example, where we've got this rise iridescence and holographic effects and dichroic, ever changing colours, depending on your viewing angle.

And that is influenced by this sort of reactive, playful, quite dynamic approach that you can have when you're experimenting with colour and pattern in your physical appearance on a digital screen and in a digital world. You could then say, well, what's happening in the physical world that can also influence the metaverse?

[00:21:36] Caroline: What you've described there about iridescent fabrics or colour changing fabrics that you may see in virtual worlds and how that's trickling into physical trends. This could be an amazing tool for innovation within fabrics and design, if people really are inspired by the trends and the colours and the otherworldly fabrics and colours and patterns and so on that they're seeing in the virtual world. If people are so inspired by that and they're wanting to bring that into physical fashion, I think we could see some really exciting innovations within textile and fashion design.

[00:22:16] We're seeing so much more rise and importance placed on the idea of personalization and customization, you could argue that's a trend in its own right and it's absolutely important for trend forecasting to think about the consumer and put them at the centre because all of this is about creating content, whether that's physical or digital, that we need and want, and will improve our experiences.

[00:23:10] Peihua: As Sally and Steve mentioned their designers and companies will under a lot of pressure with massive amount of data and strict deadlines in the fashion industries. So, In my opinion, the more data analytic skills for the future forecaster will be very variable because we could use the power of artificial intelligence that helps to push the boundary of the manufacturing design. In the metaverse the willing to express themselves among the generations that corporations might reduce the inference of the traditional like social chance on the individual shopping behaviours. Which means that the virtual variable trends might more fit into the consumer driven framework.

[00:24:39] Sally: So I completely agree with Peihua and Steve, and there is no denying that the metaverse and the advances in technology are going to play in a huge and in very important role in our future, and therefore in the way we forecast trends. I think it's really important to sort of summarize and say that with trend forecasting and questioning its validity and its need, we must remember that a lot of what we are doing, or if not all of what we're doing is, is centred around people and human centered approach is absolutely what we need to keep in, focus the power of people is really amazing and that knowledge base and, and this amazing extra information we now can gather very quickly about consumers and behaviours and people and the way we think will absolutely then influence our trend predictions. No matter what the future looks like, and no matter how quickly things are changing, we mustn't forget that people are at the centre of this as well as the planet of course. I think if we keep that in mind, we will ensure that the narratives we create are really meaningful and worthwhile.

So not just relying on the data, but actually having a human element because as you say, Ultimately it's a human industry. We have people that want to express themselves and wear things that feel right for them and reflect their values and so on. So yeah, I can see that this kind of human-centred supply chain needs to exist where we've got trend forecasters who really understand people and what's happening, and also taking inspiration, I think from those consumer trends as well. It's going be important to kind of combine them both.

Hopefully Steve, as you say, that will then make those changes that we are seeing. If we are looking at shorter lead times, we are going to have more confidence in those decisions. And actually making products that we know are going to sell and that we're not making too much and we're not making the wrong thing.Having products that are, are right for our consumers.

[00:27:46] Caroline: Yeah, collective effort hopefully to, to solve some of these issues within the fashion industry. That's fantastic. Thank you all so much. Steve, Peihua, Sally, thank you all very much for your time today. It's been a really interesting conversation and I hope that our listeners have enjoyed hearing a little bit about the changes to the forecasting industry and also the metaverse. Perhaps that's quite a new thing to some people. So hopefully you've enjoyed hearing about the future of trend forecasting and how that ties in with virtual worlds in the future. Thank you very much.