Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: Kaedim – API for 3D User Generated Content

news.ycombinator.com

Post reply on HN
Print view View original

21–30 of 41 posts

Re: Show HN: Kaedim – API for 3D User Generated Content

#22

Post by throwoutway »

> Transform your 2D art into 3D Content

One suggestion: The example on your home page, to me, looks fairly three-dimensional (as far as 2D sketches go), with shading and everything. I would suggest a better example might help with capturing potential customers.

Looks great though

Re: Show HN: Kaedim – API for 3D User Generated Content

#23

Post by tracyhenry »

pradeepb30 wrote: post #20
konstantina_ps wrote: post #14

Earlier quoted context omitted.

Hi 4thstreet, thanks for your question! Yes, we train on 2D-3D pairs. For example, we have 3D models and then we take images of them so we learn how objects look from different points of view :)

But doesn't this limit you? I mean the work here to collect the data is such a pain. While glad that you are doing this, is there any other method? Also, what are your thoughts on systems like Apple's object Capture?

It might not be as hard as you think. Get 1,000 3D models, 8 rotations for each model and 1,000 scenes. Suddenly you have 8M training pairs.

Re: Show HN: Kaedim – API for 3D User Generated Content

#24

Post by ekianjo »

konstantina_ps wrote: post #10
alex_venetidis wrote: post #9

What are the current limitations of the model which you're working on? Could you share some examples where the output is currently sub-optimal and what steps you're taking towards improving it?

Hi Alex, thanks for the question! Currently, we are not producing great results when it comes to realistic humans, animals, and vegetation. Those requests are not served as we haven't yet trained for these kind of inputs. Once our geometry reconstruction has high accuracy for hard-surface objects we'll move on to this category, collect data, train and improve.

> Once our geometry reconstruction has high accuracy for hard-surface objects we'll move on to this category, collect data, train and improve.

is that really possible though? For complex objects there should be multitude of 3d structure that fit a 2d projection and there's really no way to say "which one is the most correct one".

Re: Show HN: Kaedim – API for 3D User Generated Content

#28

Post by lasagna_coder »

From the video I gather that texturing is still a manual step? I'm a little confused how your editor showed the model without a texture, then you were able to do a perfect color fill on the different parts of the model. One of the most difficult parts of modeling is the texturing, (bump/normal map, albedo/lighting, color, etc) with lots of trade offs for how big your texture is, how much can be re-used, not to mention the actual mapping stage, which even the best "smart" auto-mapping tools do just OK.

I'm impressed by the model demo, but a lot of the time comes from determining the style (conceptual design), then implementing that style within the details (which is the baking, painting aspects of creating the model textures). You mention Robolox/Minecraft and the demo uses a kind of low poly metaverse social app, which your demo fits well, so I'm wondering who your target market is, I assume its games/apps with high volume, low detail models at the moment, is this correct?

Post reply on HN