LCC Studio

No, It's not a 360 video, It's 3d model.

@@XGRIDS the 3 d model was created from the slam k1 point cloud, right? with which software?

@@ing.adrianoflorio1403 LCC Studio. We develop the software for 3d gaussian splats training. The input is Lixel K1's scanning result.

Hi Sir，which dealer do you purchase from? Would you please send us an email to [email protected]?

Hi, you could reach out by [email protected]

Hi, would you mind leave you email address? Or reach out me by [email protected]

Hi, the visual data is not good enough for this. Lidar ensure a better result. And in this LCC solution, our scanner is a must.

Hi, moving people, animals and vehicles can be removed automatically when LCC processed while the static ones may not be. It can also generate mesh result as obj file.

You tested with their scanner or your own personal device?

Actually this was designed for space scan with depth coodinate information

Yes, it's a scanner combined lidar with visual slam

@@XGRIDS with devices like these .. why aren’t we seeing a better map .. visually on google or even Apple. Isn’t that the dream? to go anywhere and visit any place on the planet in 3D? The 3D models are atrocious on both google and Apple. I understand the data requirements.. but that’s what the cloud is for ..scan the planet every year. In 20 yrs .. it’s really like time traveling. Gaussian/nerfs are amazing visually for environments…

@@Instant_Nerf Of cours it's possible. And reconstrustion and rending in real-time in different spatial computing devices is a result that you can expect

Hey! Is this ready in the UE marketplace?

seems like a stereo camera

We use Lixel series mobile scanner

Hi sir，I'll check that. Sorry for the late reply. These days the team is busy in new product launch. You can also contact [email protected] my personal working email.

We use our self-developed 3d scanner whcih combined lidar with visual sensors.

@XGRIDS 12天前 Thank you for your interests. We do have some sample data and Unity plugin tests. If you have interests, please leave your email and some information about you and your project. You can also reach out by [email protected]

I have been fooling around with Gaussian Splats, it needs a LOT of Video memory to train, it failed on a 1080ti and that was a small dataset of just 300 photos for instance, so unless they can make it a bit more efficient, its a new tech so there's a lot to iron out.

For example to make it clear, this GTC site takes us 5min to scan and 100min to get the splats. And it's 637M. Besides images, the scanner we used combime Lidar with visual slam. So that we capture the site by mobile scanning efficiently. And the quality can be improved a lot by lidar's help. As for configuration, 3060 above is highly recommanded, and at least 2080 could be adopted.

@@TheoEngelbrechtI use the ShadowPC power tier and a program called PostShot and tho it does use a lot of its memory I've never had much issues doing anything while it was running and building. Tho it does take some models a few hours to fully clean up

Hi, we may not have a distributor in your region. You can reach out by [email protected] for the price. And you can also tell us something about your projects.

@@XGRIDS I probably couldn't afford it anyway, I'm just curious so I know what kind of threshold I'm working towards

​@@XGRIDS just give a ballpark my dude

@@mechanicalmonk2020 10k-15k USD my dude

We need the entire planet.

Hi, what's your hardware specifications?

@@XGRIDS I only have Nvidia RTX 3070 (8GB VRAM)

@@CornuDev That's good enough for LCC

Hi, we may not have a distributor in your region. You can reach out by [email protected]. Our salesman will contact you soon.

Thank you for your interests. We do have some sample data and Unity plugin tests. If you have interests, please leave your email and some information about you and your project. You can also reach out by [email protected]

@XGRIDS i have emailed you. Thank you and looking forward to a response!

Hi, we do have some sample data and Unity plugin tests. If you have interests, please leave your email and some information about you and your project. You can also reach out by [email protected]

Chunk it and blend it. At the end of the day, splats are also just point clouds. So under the hood, or manual, you can slice them up and blend them into another

SLAM，simultaneous localization and mapping. This is a 3d mobile mapping method by which you can do 3d scanning and modeling in the process of walking and moving. So that the efficiency can be significantly improved and large scale 3dgs modeling is working. Our scanner combines Lidar and visual slam to do the reconstruction. Wtih the data of Lidar, the accuracy and effects of complex scenes can be enhanced a lot as well. Hope this could help. Thank you for your comments and questions, sir~

​@@XGRIDSThank you for answering

​@@XGRIDSI don't quite understand how you get gaussian splatting just from lidar data? I thought you needed pictures also. Can you please clarify?

@@equilento hi，the scanner has built-in cameras for point cloud colonization and pictures.

@@equilento My understanding is that it takes photos for the gaussian splatting. The SLAM part is just to get accurate camera poses instead of relying on COLMAP or similar libraries for pose estimation.

Thanks so much for support