Segment 3D scene with image input by representations of segmenting 3D Gaussians with CoSSegGaussians
Segment 3D scene with image input by representations of segmenting 3D Gaussians with CoSSegGaussians
CoSSegGaussians: Compact and Swift Scene Segmenting 3D Gaussians
arXiv paper abstract https://arxiv.org/abs/2401.05925
arXiv PDF paper https://arxiv.org/pdf/2401.05925.pdf
... propose Compact and Swift Segmenting 3D Gaussians(CoSSegGaussians), a method for ... 3D-consistent scene segmentation at fast rendering speed with only RGB images input.
... method aims to achieve compact and reliable zero-shot scene segmentation swiftly by mapping fused spatial and semantically meaningful features for each Gaussian point with a shallow decoding network.
... method firstly optimizes Gaussian points' position, convariance and color attributes under the supervision of RGB images.
After Gaussian Locating, ... distill multi-scale DINO features extracted from images through unprojection to each Gaussian, which is then incorporated with spatial features from the fast point features processing network, i.e. RandLA-Net.
Then the shallow decoding MLP is applied to the multi-scale fused features to obtain compact segmentation.
... model ... zero-shot scene segmentation ... outperforms other ... on ... semantic and panoptic segmentation ... consumes ... 10% ... time compared to NeRF-based segmentation ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments