[a / b / c / d / e / f / g / gif / h / hr / k / m / o / p / r / s / t / u / v / vg / vr / w / wg] [i / ic] [r9k / s4s / vip / qa] [cm / hm / lgbt / y] [3 / aco / adv / an / asp / bant / biz / cgl / ck / co / diy / fa / fit / gd / hc / his / int / jp / lit / mlp / mu / n / news / out / po / pol / qst / sci / soc / sp / tg / toy / trv / tv / vp / wsg / wsr / x] [Settings] [Search] [Home]
Board
Settings Home
/3/ - 3DCG



Thread archived.
You cannot reply anymore.




File: Astraeus Super-Heavy.jpg (87 KB, 920x950)
87 KB
87 KB JPG
I know this would be more appropriate in /wsr/ but the experts are here.

Would it be possible for someone to make a 3D model of pic related so I can put it into tabletop simulator?

Alternatively does anyone know of someone who does commissions for this kind of thing?
>>
>>640382
ill do it for 5000 dollars.
>>
File: Sanic.png (100 KB, 405x409)
100 KB
100 KB PNG
>>640384
I'm looking at your resume, you do good work.
>>
Doesnt Games Workshop have a 360 degrees option on their site?

Just screenshot them all and throw them into meshlab :^)
>>
>>640382
hehe I'll do it. I'm huge 40k fan
and also few months back, I have made Gaunt ))
if you want I'll give him too (on top). actually I made him to just give away to 40K community, i'm too lazy to prepare 3d printable version..
>>
File: ezgif-2-3c2a7d90ce.gif (2.21 MB, 600x409)
2.21 MB
2.21 MB GIF
>>640382
>>
>>640406
Thats actually really good compared to some of the stuff on the workshop.

>fan fricking tastic considering the limitations of forgeworlds 3d viewer.
>>
>>640385
funny thing is, I would actually charge around 5k to model that vehicle in the detail that it shows in the image.
>>
>>640495
In what world would someone pay you 5k for a job like this when I can get the same from Pajeet or some Chinese Zergling for less ?
>>
>>640495
>5k
lol what the fuck
>>
>>640498
>>640504
You guys are scaring me. Are you just millennials that dont know what to charge or are you clueless?

A few weeks ago I was hired to model something for a feature film that was a common household object. They needed it to be high detailed and it took over a week and I got about 5k. If I got 5k then the studio had to pay for texturing, and lighting, and animation so imagine the real cost? easily 30k. Then add on the profit the studio wants.
>>
>>640506
Depends. It’s much easier to model something out of figure photos than from a concept for example. And then there are the amount of details, a miniature doesn’t need so much.
>>
>>640506
30+ years of doing CGI here...and I would never pay my employees 5k for a job like this, because I need to pocket all the profit to fuel my extravagant Boomer lifestyle.
>>
>>640506
May I ask what kind of object it was? I'm still trying to gauge how much to ask for what.
>>
>>640495
Considering that gif up there just needs the white crumpled up bits deleted. and He did that in an hour (an hour exactly now that I look at it) You are obviously talking about building it from scratch.
>>
>>640518
maybe it escaped your notice but it has no underside. Once you've deleted the crumpled bit it's gonna look like swiss cheese, too.
>>
File: ezgif-2-4816caad12.gif (2.15 MB, 600x419)
2.15 MB
2.15 MB GIF
>>640382
>>640406 2nd test with the Astraeus tank
program
https://github.com/alicevision/meshroom
images
https://imgur.com/a/LewpMQT
Output without clean and without mirror
https://ufile.io/itd2k
>>
File: ezgif-2-29981ffc56.gif (2.08 MB, 401x253)
2.08 MB
2.08 MB GIF
>>640545
last test, I just expanded the images to 4800x4800 and select ultra in the FeatureExtraction node, 25 pictures, 3 of them failed, arround 90 mins in my old I5, no need to mirror this time
the file without clean, just the output from meshroom
https://ufile.io/koqp0
>>
>>640532
For the purposes of Tabletop Simulator, he doesn't need the bottom.
>>
>>640382
>tabletop simulator

Are you using a 3D game engine? Pre-render or real-time ingame?
>>
>>640645
I'm sure OP is talking about using steam workshop for tabletop simulator https://www.youtube.com/watch?v=gc-VC2X8bZ0
>>
File: (1).gif (2.86 MB, 600x383)
2.86 MB
2.86 MB GIF
I've just finished the Retopo, the bake was fucked and I had to paint a few, the texture need more work but the model looks "good" and I'm tired of this, arround 2k tris
>>
>>640675
Looks fantastic.
>>
>>640728
Real nice.
>>
>>640728
Just imported it into TTS, and scaled it up to the proper size (it was hilarious it spawned in at like 1cm in length). Do you have a pseudonym you use so I can give you proper credit on the workshop page, or do you want to do that yourself?
>>
File: Capture.jpg (141 KB, 1084x866)
141 KB
141 KB JPG
>>640731
>>
File: 1.png (1.49 MB, 1920x1080)
1.49 MB
1.49 MB PNG
>>640731
No need, was nice practice, reupload i deleted accidentally
https://ufile.io/eb3l7
>>
>>640564
dude this is sick!! been trying to replicate your results. I'm getting most of my pics failing, then it will only generate a small piece of the Mesh. Did you scrape the photos right off the games workshop website? with their 360 view I can only get like 19 or so diff angles, at like 800x800 resolution. I also tried uprezzing to 4800x4800 and removed backgrounds in PS but no luck, only will get like a corner part of the Mesh to generate. I also tried Ultra setting in tha Feature Extraction node but no luck. Do you have any other protips/ workflow suggestions ?? This is epic would love to be able to do something similar, maybe even scrape stuff from pixelsquid/turbosquid. thx senpai for the inspiration and keep up tha great work!
>>
File: Warhammer.png (1.32 MB, 1297x815)
1.32 MB
1.32 MB PNG
I haven't try to remove the backgrounds, uprezzing to at least 3kx3k work for me, increase the contras in the pictures and change the camera field of view from 45 to arround 35 work also sometimes, I sometimes enable sift float and sift upright in feature extraction, feature matching and structure formation, I've only tried with the games workshop pictures, some of them work great and some other don't, the Hierophant was a pain in the ass and the result horrible, to extract the pictures just enable rotation and open the image in a new tab, after that increase the last number,some of them has diferent levels you have to change that too, when that happens I usually get 2 lvs and 1 picture from the top, if the result is bad you could take more, I had 2 results even worst than the Hierophant with a couple of basic space marines.
>>
>>640495
Nah, get the fuck outta here, any mediocre modeler would do this shit in 1 week tops. I probably can model this shit in 2-3 days. That is at most 2k for a senior modeler salary. Stop lying to people you faggot
>>
File: 1536217736219.jpg (327 KB, 700x1690)
327 KB
327 KB JPG
>>641258
thanks you so much for getting back to me!! I will retry with those suggestions. Wicked appreciate it. Was having trouble finding documentation for meshroom online. May the Emperor protect you in all your battles (and projects!)
>>
>>641264
I've just found this, I hope it helps
https://www.youtube.com/watch?v=k4NTf0hMjtY&app=desktop
>>
>>641312
ah ty my g! i ran thru the tut, good overall review of getting started, i just need to figure out what all the settings with nodes actually do lol. For that ill probably need to study more of the principals of photogrammetry. Getting some decent results now tho to start. good lookin out!
>>
>>640545
>>640675
Looks amazing.
Are there any tutorials?

Could I use photos of myself and make a model of myself as well?
>>
>>641312
Nice thanks.
>>
>>641405
>Could I use photos of myself and make a model of myself as well?
Sure, but the program will ask for a couple of things from the camera, you can also grab frames from turnaround videos from toys scale models and things like that from youtube, there's toons of them
>>
>>641422
>will ask for a couple of things from the camera
Like what?

>you can also grab frames from turnaround videos from toys scale models and things like that from youtube, there's toons of them
say, could I just take a video of an object instead of photos?
What would be the better option?

I'm thinking recording an object with video would be easier and there'd be more frames as well.
Can the program take video files or would I have to export the frames to images somehow?
>>
>>641426
field of view I think and maybe model
>What would be the better option?
I use frames from videos if aren't blured and works great, but you have to take frames, the program doesn't support video, I suppose that take toons of pictures is a better option but is faster get a video and extract images from there, you usually have to uprezzing that images but that isn't a big deal, the only thing is that the images can't be blured
>>
>>641426
I use arround 200 images / model
to extract the images you can use blender and make an image sequence
https://www.youtube.com/watch?v=SJdoPfkYDuw [Embed]
and change the frame Step to the number that you need and you can resize them there too, that's all
>>
>>641428
>the only thing is that the images can't be blured
Wouldn't upsizing do that?

If you don't want blurry images have a high "f" stop and manual focus.

Also I'm sure there's a scripting/software solution to finding blurred images though.
>>
>>641435
you don't need a high definition picture, you need a big picture, I use 3800x2100, if it isn't blured in 720p it is ok, less than 720x560 will have less detail and a worse result but you can try, just resize that to arround that
>>
>>641444
Why do I need a bigger picture though?
What's makes it better when it's upscaled?

I think Higher definition is better though.
>>
Wow didn't expect this request to turn into a teaching thread?
>>
>>641470
>Why do I need a bigger picture though?
by default the program compute depth maps from images that are 2x downscaled. You can also change the scaling value in the depth maps computation settings. If you don't want to mess with the program is better upscale them





Delete Post: [File Only] Style:
[Disable Mobile View / Use Desktop Site]

[Enable Mobile View / Use Mobile Site]

All trademarks and copyrights on this page are owned by their respective parties. Images uploaded are the responsibility of the Poster. Comments are owned by the Poster.