How fashion startups get accepted into tech accelerators | Hacker News Reader

How fashion startups get accepted into tech accelerators

voguebusiness.com

41 pointspshaw11 comments

I'm surprised I haven't seen more fashion/clothing startups (maybe I have just missed them?).

Specifically I see great potential in just-in-time production of garments using 3D scanning, motion tracking, "curve" fitting algorithms and automatic manufacturing. I would LOVE to be able to walk into a store and buy a pair of jeans that actually had a great fit. Would definitely pay premium. Tailors doesn't scale well.

Unfortunately this is probably further than many in tech realise. The tech is getting better but the manufacturing tolerances on most clothing is far too far off.

Typical manufacturing tolerances on "high street" clothing (not high end) will be about half a size. On budget clothing it could be as much as a whole size out between batches.

It's not until you get to much higher end than most people are willing to pay that you get to the good enough tolerances. a £50 shirt (which is pretty high end for most people) will still be out by a lot. A £150 shirt might have the potential to be close enough on the tolerances, but at that price point sizing isn't the factor that people care about, it's luxury brand names.

I'd be happy to pay a premium for good fits, but the premium I'd be willing to pay (and that I bet many/most would be willing to pay) is in the 10-30% range, not the 200-300% range that it would likely need to be.

One free trick to getting a better fit is to try on more than one of the “right” size for the same article of clothing.

Free business idea: Don’t worry about loose manufacturing tolerances, but make a way of precisely measuring each individual finished garment, then match the specific garment to the specific customer.

I like it.

Personal AR devices with 3D cameras will make it easy for someone to get precise measurements and re-scan after any weight change. On the manufacturing side, I think we need to combine cameras with clothing material data and robots that stretch the clothing in various directions to ascertain actual size. The output measurements could be matched up with the 3D scans of customers to make accurate size recommendations.

In addition, the manufacturer's measurements could be turned into a 3D model of the clothing that can be "worn" by a deepfake digital version of the customer. CGI footage would be generated of Deepfake Customer inhabiting a curated branded setting, showing how the clothing would fit on the customer from multiple angles. This wouldn't only be for the customer's viewing. An ML Aesthete would view Deepfake Customer catwalking in tens of thousands of permutations of the store's clothing, recommending the combination of merchandise that would be most attractive, adjusting for the customer's preferences.

If the customer is still undecided between several articles of clothing, photos from the deepfake catwalk could be posted on a website like Photofeeler.com, where human opinions would be rendered in minutes and compared with the customer's baseline attractiveness to compute an "attractiveness boost" score. ML Aesthete would be "punished" for differences between their score and the human scores, until their scores match up and human opinions are no longer necessary. Alternatively, the customer can post an Instagram story with a voting mechanism. If anyone with ML and/or robotics experience would like to work with me (I'm currently non-technical but have business acumen) on the whole system described above, even if perhaps as an aspirational side project, please reply below.

> at that price point sizing isn't the factor that people care about, it's luxury brand names.

Is this your personal experience or do you have data on this? YSL is one example has very tight (thin) fits on their jackets. You will buy the one that fits the best, otherwise it's gonna look ridiculous.

I think the OP's point is that at that end of the market, the salesperson measuring you up for the custom fit/alteration of shirts is still good enough [and pays for themself] to get dimensions you'll be happy with, and 3D scanning tech is a niche pitch.

Yep this was my point. It's not that people don't want things that fit, it's that this is not the primary motivating factor in that part of the market.

What’s to explain? Consumer tech is fashion products, not tools.

Or have you not noticed whenever you first start dating someone new your choices in app brands elicits the same kind of responses as your choice in clothing?

Key takeaways:

Fashion companies are looking to invest in “enabling” startups that help them implement new business models or technologies.

Venture capitalists and incubator executives like founders who don’t gloss over the challenges they face.

Founders should have concrete data on the competition and potential consumers, as well as preliminary results of having tested their idea.

Reply on news.ycombinator.com