taichi_3d_gaussian_splatting icon indicating copy to clipboard operation
taichi_3d_gaussian_splatting copied to clipboard

An unofficial implementation of paper 3D Gaussian Splatting for Real-Time Radiance Field Rendering by taichi lang.

Results 37 taichi_3d_gaussian_splatting issues
Sort by recently updated
recently updated
newest added

Hello, I've seen references to KITTI in both the "config" and "tools". I wanted to ask if you've tested the KITTI dataset and if you can provide the test data.Thanks!

I got wrong color between taichi viewer and PLY which converted from parquet. It got right color when opened by viewer but got strange color with PLY which is more...

Is Enable camera pose optimization #83 solved? Thx!

When I trained the model with 8G-1070Ti and 24G-3090Ti, the speed increase was not significant. Compared the original-version, the taichi-version is 2-3 times slower, is it possible to speed up...

Hi there, Thanks for your great piece of work - I'm keen to try simplifying a few things and interested in your opinion: First one is relatively simple: There's lots...

![2023-11-23_22-20](https://github.com/wanmeihuali/taichi_3d_gaussian_splatting/assets/62792139/f7311976-6f39-4bed-a782-214670b259ba) ![2023-11-23_22-21](https://github.com/wanmeihuali/taichi_3d_gaussian_splatting/assets/62792139/2581a0a0-6a3f-49b5-8317-2a0a666871a5) Reconstruction fails after iter 500, how should I adjust the parameters to fit my dataset.

I am new to gaussian splatting and taichi as well. Is there quick way to speed up training process without loosing quality. Is it possible to do batch processing? P.S....

thank you for the amazing code! just got it training today in colab. i would like to try implementing batching. i saw it listed in the engineering TODOs. did you...

Is there a timeline for adding metal backend to the Taichi training code? I have other GPUs but apple silicon macs have a lot of unified memory and are very...

Hi, Thank you so much for the great work! I was wondering if you have also tried the NeRF-synthetic dataset, where the Gaussians are initialized randomly (in the original paper)....