OpenSplat
OpenSplat copied to clipboard
pierotofy
Production-grade 3D gaussian splatting with CPU/GPU support for Windows, Mac and Linux 🚀
There should be a way to reduce/filter floaters.  These significantly reduce the quality of visualization. I thought about doing a nearest neighbor search from of the splat means compared...
Vulkan has compute capabilities and could be used for speeding up the training with other GPU brands. https://www.khronos.org/blog/getting-started-with-vulkan-compute-acceleration It could also be possible to compile torch to use Vulkan, but...
I got this out of gdb, always happens during images: Thread 1 "opensplat" received signal SIGSEGV, Segmentation fault. __memmove_evex_unaligned_erms () at ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S:317 317 ../sysdeps/x86_64/multiarch/memmove-vec-unaligned-erms.S: No such file or directory.
Hello! Working through the installation on a CUDA machine and WSL/Ubuntu 2204.3 on WIndows 11 and getting tripped up during the cmake command in the readme. I followed the Build...
There are several parts that could be optimized: * Implement multi-threading * Benchmark and eventually port the project/compute_sh functions from libtorch to pure C++ (will need to manually write the...
Current 3DGS training pipeline is heavily relying on the SfM initialization, which introduces a significant overhead when scaling to large scenes. Technically, 3DGS should be capable of producing relevantly good...
Ubuntu 22.04 container in podman (Fedora host) ROCm version amdgpu-install_5.7.50703-1_all.de libtorch version libtorch-cxx11-abi-shared-with-deps-2.2.1+rocm5.7.zip rocminfo reports CPU at 0 and GPU at 1 ``` ******* Agent 2 ******* Name: gfx1103 Uuid:...
Hello, and congratulations on your awesome work! Now that you've added Colmap support, I'm really excited to dive in. Question, do you plan to support 2d image masks? These would...
Today, I have read discord that gsplat & nerfstudio has a big bugs. After it fixed, it being said that the quality is increasing. The bug is about The calculation...
