Code implementation of CVPR2025 paper "FreeTimeGS: Free Gaussians at Anytime and Anywhere for Dynamic Scene Reconstruction"

[jcwang-gh]

Thanks for the inspiring work. I wonder if you have any plans to open-source the code of the CVPR2025 paper "FreeTimeGS: Free Gaussians at Anytime and Anywhere for Dynamic Scene Reconstruction"?

[CBY-9527]

I have the same interest as well.

[maturk]

Same here, would be great to get an update.

[zihanzheng-sjtu]

I also hope to get the code update.

[wusar]

I have the same question. As indicated on the official webpage of the 1st Volumetric Video Challenge at SIGGRAPH Asia 2025 (available at https://www.4dv.ai/research/sig-asia2025-volumetric-video-challenges#challenges), this paper serves as the baseline for the competition. Therefore, I also hope to obtain the implementation materials of this paper, which will enable me to run and replicate the baseline results.

[GavinSun0921]

Same here, would be great to get an update.

[sipkode]

Since the authors of FTGS don't seem to be very forthcoming with the code, y'all can check out OMG4 which does have code, and which claims even better compression!

[vicksEmmanuel]

Since the authors of FTGS don't seem to be very forthcoming with the code, y'all can check out OMG4 which does have code, and which claims even better compression!

Takes 2+ hours

Training progress: 6%| | 1670/30000 [09:00<2:19:09, 3.39it/s, Loss=0.1466