Is anyone using 3D Body Scan Hardware or Apps


#1

Hello, I know that there are plans in the works for Seamly to become or have the capacity to use for actual 3D body scans and resulting measurements. But until that is fully developed, I am curious if anyone has been using any other solution in the interim for getting those exact measurements.

Currently, I hand measure clients and hand input measurements into Seamly (and I love using Seamly2D for making the patterns). But I would love the ability to automate the measuring and data entry. That way I wouldn’t have to meet directly with a client (and I could even even work with clients out of state or in another country).

I see a lot of 3D scanning solutions out there including iPhone Apps and iPad attachments. Open Source Raspberry Pi cameras. And within the fitness category I also see that there is a lot of promising ideas related to body shape tracking (almost like a replacement for a home scale). There’s even some solutions that use the Microsoft XBox Kinect for taking 3D renderings of the body.

Perhaps one of these solutions can be used to gain the measurements necessary for making custom fit patterns. Is anyone currently melding any of these solutions or services with Seamly?


#2

part of the issue is deciding which measurements you will need. eventually, “all” scan sources could capture “all” measurements. in practice, that is just not the case. can you describe what patternmaking system you use or what set of measurements you need to capture?


#3

I went to a health club and had a scan. The scan was provided by a franchise, FIT3d. I got a 3d image that contains a really accurate image of my body and which I can manipulate in blender. The only measurements that were provided automatically in the FIT3d information are all circumference. They include Neck, Bust, Waist, Hips, Left Biceps, Right Biceps, Left Forearm, Right Forearm, Left Thigh, Right Thigh, Left Calf, Right Calf. The scan provided no vertical distance information, not even height.

So I hope that helps explain my concern.

I do not mean to be discouraging. Getting the measurements into the seamly software is a trivial technical issue. Capturing the measurements from the raw scan data is key and would be possible when the person(s) with the expertise to do that can be identified and agree to work with us.


#4

Oh you’re not being discouraging. I appreciate the insight. I’m genuinely interested in seeing Seamly extended. Just wish I could program (but Seamly may be the impetus I need to learn how).

I thought maybe that a solution like what you described (like FIT3d) may be a way forward. But ultimately there are some measurements that you’re still left wanting (and still ultimately need to format and enter manually). Vertical measurements of coarse being as important as circumference.


#5

I truly think a good first start would be to define a subset of measurements that are useful. Do you have an opinion about what that would be? (starting with the measurement already recognized in the SeamlyMe program) Please see Measurements - Seamly2D

If you are willing, one thing that you could do to help the Seamly2D project get closer to the goal of 3D is to help write a specification. If that sounds like too big a task to you, I have a simple task and would like to ask for your help. Can you review the information from the reference provided above and propose a list of measurement that would be sufficient to start with. Eventually, some programmer is going to have the task to extract the measurements from scans and it will be most helpful to have a short list of measurements as a start. It would be a huge help if you can do the first draft of such a list. Please identify what patternmaking system you use and tell why you chose that particular list.


#6

Yeah. I believe I could help with this.

I currently adhere mostly to a personal group of measurements that I have assembled using a mash of Lori A. Knowles “Patternmaking for Fashion Designers” and some current ASTM measurements. There’s obviously some differentiation, however many of the measurement names can be cross referenced (although the values and exact names are not always the same).

Ultimately, I found Lori A. Knowles Pattern making books to be very helpful when learning pattern drafting, and it’s suited me well over the years. I’ve created my own workflow that diverges slightly from her methods, but overall her system is what I like to use.

I do design for production though, so for my most recent measurement file(s) I used mostly ASTM measurement values plugged into existing Seamly names. And then for custom clients, I alter this file and enter the clients individual measurements (as opposed to using the ASTM value).

I don’t really have an issue with most of the names for measurements found in Seamly. But as you stated there is no set standard, so it can be confusing to see different names for the same value (or similar names with different ways of coming about at the value).

Although some may say ASTM measurements are limited, it’s still a fairly defined list of common measurement points to use as the main specification.


#7

I want to ask @slspencer to weigh in here, because she is the one who has painstakingly scoured through many different patternmaking systems and she know far more than I about different names for the same value. One thing that she and I have discussed is the need to create templates for each patternmaking system so that someone using Seamly2D and SeamlyMe would not have to learn names other than what is documented in the book for “their” preferred system.

At any rate, if you can come up with the list of measurements that you need to be the useful minimum startup set and map those measurements to the names documented at Measurements - Seamly2D, that would be a huge help.

Please see what I have documented as a starting point for a template for using the McGunn patternmaking method. McCunn - Seamly2D


#8

I’ve done a similar thing for the Aldrich system on PDF, if you’d like it.


#9

Something that just popped into my head is to create a ‘garment’ (based on the desired pattern system) for the created body model that simply consists of the measurement lines that system uses. Allow the user to move the measurement nodes around the model, disabling the ones that aren’t needed, then output a list of the resulting measurements. (One such ‘garment’ should contain all of the measurements that Seamly2d supports.)

I’m not sure just how difficult it would be to code it, but I can see its usage being intuitive.


#10

@KeithFromCanada, that is a really interesting idea. From a user point of view I love it. From a coder point of view I need to point out that extracting measurements from a body scan is a non-trivial task and I know of no software that does it yet. Even commercial software. I say that in response to your suggestion or using “all” of the measurements.


#11

Hmmm… While I am operating in total ignorance of the specifics, it seems reasonable to me that you could define initial measurements as two lines defined by three points on a plane set at a particular position and rotation on the model. (Circumferences would just be four lines on the same plane.)

To ‘drape’ the measurement over the model,

  1. The endpoints are be moved on a line parallel to the surface normal of the model until they are each resting on the surface.
  2. The centre point of each line is made an endpoint, creating two lines from each one.
  3. Repeat steps 1 & 2 until there are no more points that can be moved more a set value. (e.g. 2mm)
  4. The measurement will be the sum of the lengths of the line segments.

Does that sound reasonable?


#12

sounds quite reasonable as an approach. the issue is that there is nobody AT THIS MOMENT IN TIME associated with this project who has the time to implement this. I don’t mean to shut you down. I simply want you to understand why you are not likely to see progress on this great idea immediately


#13

Fair enough. I’m just providing my thoughts on the matter, not demanding that someone do something about them. If someone does come along with the time and willingness to take it on, perhaps my ideas could be of help.

One option to help spur things (in general) along would be to post bounties on the features that are most wanted, backed by a small crowdfunding campaign asking users to vote with their wallets on which features they most want to see. There are literally tens of thousands of coders on this planet who would be willing to code any features needed even if they weren’t paid much by our standards. What this project might need more than anything is someone to hunt down funding to push it ahead.


#14

Actually, yes :slight_smile: I use Nettelo. It is designed for taking tailor’s measurements. I’m not sure if they have an android version; I’ve used this on my iPhone for a couple of years. When they first started, you were allowed to use the 3D model it creates (it would email you the obj file) along with the measurements free of charge but now that is a pay per month perk. It does still give really accurate results. If the directions are followed closely, it is accurate for nearly every measurement. Even when it’s wrong, it isn’t usually more than a half inch off of the measurements I take by hand. The only exception is that of my underbust measurement (off by about 2 inches), but I’m busty and pictures can only do so much. The measurements it gives are almost all of what you would need and you can pick points on the 3d figure to measure, but I haven’t ever used it so I can’t tell you how accurate it is.


#15

Oh, wow! So there is something that does this :slight_smile: I found a number for ‘surveys’ on the internet last year that students did comparing the measurements of children in different countries to the standard measurements of garments sold in that country and they were using some sort of scanner to record a few different body measurements of hundreds of children. Perhaps it was something like this that they used.


#16

Actually, yes :slight_smile: I use Nettelo. It is designed for taking tailor’s measurements. I’m not sure if they have an android version; I’ve used this on my iPhone for a couple of years. When they first started, you were allowed to use the 3D model it creates (it would email you the obj file) along with the measurements free of charge but now that is a pay per month perk. It does still give really accurate results. If the directions are followed closely, it is accurate for nearly every measurement. Even when it’s wrong, it isn’t usually more than a half inch off of the measurements I take by hand. The only exception is that of my underbust measurement (off by about 2 inches), but I’m busty and pictures can only do so much. The measurements it gives are almost all of what you would need and you can pick points on the 3d figure to measure, but I haven’t ever used it so I can’t tell you how accurate it is.

@slspencer it sounds like it is worth taking a look at Nettelo to see what it does


#17

Whoa! That’s pretty sweet! Too bad they haven’t done an Android version.


#18

You know… a project with far more requirements received bids that averaged around $5500 on FreeLancer.com. That amount isn’t so sky-high that perhaps sponsors/donors/crowdfunding could pay for adding the ability to take measurements as Nettelo does. I’ll bet a lot of people would be willing to contribute on Patreon/GoFUndMe/etc. with that goal.


#19

FWIW: I have gotten excellent results using the Morris book, Women’s Garment Cutting and Making with the AASTM measurements. There are a couple that I’ve had to tweak when I am using Parisian Ladies Tailoring System but I really rely on Morris. Like I said, I’m busty and trying to get the more modern approaches to pattern making to work for me is a nightmare. The neck to bust point was the bane of my existence until I found Morris.

I got so much more out of Seamly when I setup my formulas to match the book descriptions-- I took the time to create my own version of the AASTM measurements that someone else had posted on here and now I have a whole range of sizes, I can input anyone’s measurements into the same template, and I only have to make the pattern once.

If you’re really into tech working together and you can afford Marvelous Designer, you should look into Clo. It’s the MD platform with a few extra bells and whistles that are for designers. All of the things that Seamly can’t do just yet-- matching curves, custom seam allowances (I really don’t need an extra quarter inch on my fold line, thanks), save to DXF files, etc. I usually export to Clo, make edits, work out how those edits would translate in Seamly, repeat the process until I worked out a healthy chunk of issues, then put the final result in Illustrator to print with page numbers and little icons to match up pattern pieces. That way i waste far less ink and fabric :slight_smile:


#20

You can do this. Have a look at this post

custom seam allowance