Post-training semantic lifting for 3D Gaussian Splatting, with an evaluation that separates detector, lifting and representation error on Replica and ScanNet++. Bachelor's Thesis repository
-
Updated
Oct 9, 2026 - Python
Post-training semantic lifting for 3D Gaussian Splatting, with an evaluation that separates detector, lifting and representation error on Replica and ScanNet++. Bachelor's Thesis repository
3D Gaussian Splatting on ScanNet++ Data: Best Practices and Workflow Guide. In this way splat coordinate will match the coordinate system with the Scannet++ coordinate. Same workflow for Triangle Splatting+, MeshSplatting
A unified, feed-forward model integrating the MapAnything 3D reconstruction backbone with Mask2Former for multi-view consistent panoptic segmentation. This approach provides feature-rich and metric accurate semantic maps for indoor robot navigation while eliminating the need for posed inputs and costly test-time optimization.
PIVOT - film a room, ask if furniture fits: SAM 3.1 + Muse agent with verified 3D geometry. 0/36 false claims vs 15/36 for Muse alone. Meta Global AI Hackathon 2026
To associate your repository with the scannetpp topic, visit your repo's landing page and select "manage topics."