Thanks for your work!
In the paper we see such a sentence，“Gaussian semantic tracing is achieved by unprojecting 2D segmentation masks into 3D Gaussians and assigning each Gaussian a semantic tag”. Does this indicate that multi-view masks are used to initialize the semantic information of each Gaussian point,Specifically, the weight of each semantic.
another question：Where is ”Gaussian tracing algorithm“ reflected in the code?
Thanks for your work! In the paper we see such a sentence，“Gaussian semantic tracing is achieved by unprojecting 2D segmentation masks into 3D Gaussians and assigning each Gaussian a semantic tag”. Does this indicate that multi-view masks are used to initialize the semantic information of each Gaussian point,Specifically, the weight of each semantic. another question：Where is ”Gaussian tracing algorithm“ reflected in the code?