search

agelosk / dynmf

(ECCV '24) DynMF: Neural Motion Factorization for Real-time Dynamic View Synthesis with 3D Gaussian Splatting
67 stars 1 forks source link

question #2

Open hhxphh opened 2 months ago

hhxphh commented 2 months ago

Dear authors, what wj mean? I dont see an explanation in paper? bj(t)=MLP(t/T;wj)

agelosk commented 2 months ago

w_j are the weights of the MLP. Each trajectory basis has its own learned weights.

hhxphh commented 1 month ago

w_j are the weights of the MLP. Each trajectory basis has its own learned weights.

thanks ! man!

hhxphh commented 1 month ago

w_j are the weights of the MLP. Each trajectory basis has its own learned weights.

when will the code come? I miss it everyday, I cant live without your code