Segmentation of 3D Gaussians using Masked Gradients
SessionPosters Presentation
DescriptionA novel method for segmenting 3D Gaussian splatting models using 2D masks and inference-time gradient backpropagation, offering improved accuracy for applications in augmented reality and 3DGS editing.
Event Type
Poster
TimeThursday, 5 December 20241:00pm - 2:00pm JST
LocationLobby Gallery (1) & (2), G Block, Level B1
TE
EH
