Researchers from China have developed a new AI technology called 4K4D, which is a 4D point cloud representation that supports hardware rasterization and enables unprecedented rendering speed. This technology aims to tackle the issue of slow rendering speed when it comes to real-time view synthesis of dynamic 3D scenes at 4K resolution.
https://zju3dv.github.io/4k4d/
https://github.com/zju3dv/4K4D
(code is coming soon)
4K4D: Real-Time 4D View Synthesis at 4K Resolution
Paper | Project Page | arXiv | Video (Coming Soon)
Repository for our paper 4K4D: Real-Time 4D View Synthesis at 4K Resolution.
Source code is coming soon!
For more high-resolution results and more real-time demos, please visit our project page.
Discussion (1)
🙏Thanks for the Info and Post.
Sounds interesting, but as far as I understand it will certainly take some time for software manufacturers to implement these codes (sources) and adapt their programs. Or is there already end-application software where it is used? The picture for example. shows some kind of software but is that so laboratory stuff that is simply used for development.
And I can't quite place the word "view" either. So the view window is supposed to be used primarily for viewers in programs, and end result rendering for main outputs is still to render timeconsuming?
I probably don't understand some things very well hear.🤔
Greetings to you Vdmo & the Kangerus, bennoH.🐼🙋 from mountainous snowy Switzerland