- 59
- 27 799
Visualito Com
เข้าร่วมเมื่อ 4 มิ.ย. 2015
วีดีโอ
3D Asset Generation With NVIDIA Edify 3D
มุมมอง 1Kหลายเดือนก่อน
build.nvidia.com/shutterstock/edify-3d
4D Reconstruction with MonST3R
มุมมอง 932 หลายเดือนก่อน
github.com/Junyi42/monst3r monst3r-project.github.io/
Gaussian Splats from Image Pairs
มุมมอง 992 หลายเดือนก่อน
Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairs github.com/btsmart/splatt3r
3D Video with DepthCrafter
มุมมอง 1332 หลายเดือนก่อน
DepthCrafter: Generating Consistent Long Depth Sequences for Open-world Videos github.com/Tencent/DepthCrafter
3D Video with Depth Pro
มุมมอง 1672 หลายเดือนก่อน
Depth Pro: Sharp Monocular Metric Depth in Less Than a Second. github.com/apple/ml-depth-pro
Gaussian Splatting with MASt3R-SfM
มุมมอง 1.6K2 หลายเดือนก่อน
3D Reconstruction from Historical Aerial Video Footage, Reichstag Damaged by Air Raids, 10 July 1945, Berlin Reichstag Dome demolished with the bombings of Berlin through World War II github.com/naver/mast3r/tree/mast3r_sfm github.com/hbb1/2d-gaussian-splatting
Image to Video with Gen-3 Alpha
มุมมอง 693 หลายเดือนก่อน
Experiments with Runway Gen-3 Alpha image2video, video2video
Single Image to 3D Video with ViewCrafter
มุมมอง 1353 หลายเดือนก่อน
arxiv.org/abs/2409.02048v1 github.com/Drexubery/ViewCrafter/
Video Segmentation with Segment Anything Model 2
มุมมอง 4715 หลายเดือนก่อน
Video Segmentation with Meta Segment Anything Model 2 (SAM 2) ai.meta.com/sam2/
Single Image to Point Cloud with Depth Anything
มุมมอง 3735 หลายเดือนก่อน
github.com/DepthAnything/Depth-Anything-V2
Single Image to 3D Point Cloud with Depth Anything V2
มุมมอง 1.3K6 หลายเดือนก่อน
github.com/DepthAnything/Depth-Anything-V2
Reichstag 3D Reconstruction from Historical Aerial Video Footage
มุมมอง 2227 หลายเดือนก่อน
3D Reconstruction from Historical Aerial Video Footage with Gaussian Splatting Reichstag Damaged by Air Raids, 10 July 1945, Berlin niujinshuchong.github.io/gaussian-opacity-fields/
3D Reconstruction from 80 years old video
มุมมอง 2.2K7 หลายเดือนก่อน
3D Reconstruction from Historical Aerial Video Footage with Gaussian Splatting Locomotive Depot Damaged by Air Raids, 10 July 1945, Berlin niujinshuchong.github.io/gaussian-opacity-fields/
Surface Reconstruction with Gaussian Splatting
มุมมอง 5967 หลายเดือนก่อน
Surface Reconstruction with Gaussian Splatting
Is it possible to share the code for converting depth to 3D?
Guys, it is quicker to make a Gaussian splat and then export it as a mesh or direct photogrammetry? I'm new to the subject, so any help would be appreciated. Btw, i use SuGaR to export the mesh, the quality isn't so great but works 60% of the time.
This looks great! How does it compare to hloc? I'll probably give them both a try later, but i didn't like the hloc workflow first time i tried it.
Salut.. j'aimerais bien savoir comment faire l'animation 3d avec NVIDIA edefy sur andr😢
This looks cool, but the surface seems to be unusable...
Do you know if it's possible to do passthrough with this? i.e. the gaussian splat with passthrough video outside of the model?
Pls download
That's great, I've been wondering lately with the SoTA is still COlmap given how old it is. Did you implement it directly from the repo, or do you know if there is a more 'user-friendly' implementation ?
Yes, I used the implementation directly from the repository github.com/naver/mast3r/tree/mast3r_sfm
I am so impressed by this technology. I've been testing a lot with my own drone shots and postshots for a few days now. A little tip on the side. Today I found a workflow for cleaning up splats in a video by Olli Huttunen (Let's Clean the 3D Gaussian Splatting models and remove floating artifacts!) that is really good. Everything then runs via Unity. We live in really exciting times, with endless possibilities =)
What video 2 video tech you used?
runwayml.com
Perfect!
I’d love to know how to view Gaussian splats via meta quest 3 headset to create something that could be used where I work.
Woah, amazing
BRAVO!!
Which is the best depth camera? What parameters to check for a depth camera?
Wow, can i upload own Gaussian thorough luma ai?
crazy :O
Can you share your PCD Transform code ?
What software/workflow did you use?
1. Input image 2. Image => Depth image with Depth-Anything-V2 - github.com/DepthAnything/Depth-Anything-V2 3. Conversion "Depth image => 3D point cloud", my code 4. View point cloud in MeshLab
This is exactly the type of stuff I‘m interested in seeing more of. Taking older footage and give life to it in a new way. Maybe in some cases one could be able to rebuild for example an old village that has been filmed from various angles in the past into a 3d model. Really cool stuff. Great work!
Yeah, same as you
Tutorial please
Very Impressive!
This is amazing.
Amazing! How many images did you use?
76 images
Is that a 3d mesh at the end? Amazing work btw
Yes, it is 3D mesh extracted from 3D Gaussian Splatting
@@visualitoXrmay i ask how this is done?
niujinshuchong.github.io/gaussian-opacity-fields/
Please tutorial
Have you managed to make stereoscopic gaussian splats?
*Promosm* 😄
in quest 2 is a bit laggy, wich headset you are using ? thanks!
Quest 3
Nice build!❤❤❤
This is promissing. How many frames per second? Could a quest3 game be made with gaussian splatting meshes!
Frame rate is between 15 and 70 FPS. I'm afraid it's too low for gaming
Hello, this looks amazing!! Can you make a tutorial for how you do it in AR?
Kdyby alespoň ty názvy zastávek byly správně... Šimkova není pro MHD a Fortna už neexistuje asi 30 let.
whats the name of the software on the tango?
Slušná práce, doufejme, že se Vltava nezvedne o těch 30m :D.
Hi, can I ask you, which app are you using, for viewing just depth sensing camera input? Thanks a lot :)
Hi, I use my application for that
Such high quality texture..
Is this from the Lenovo Phab 2?
Yes, Lenovo Phab 2 Pro