cvg/nice-slam
dense SLAM with neural implicit encoding for indoor scenes
View on index · View in 3D Map
Related repos
- Visual SLAM using a 3D prior for dynamic environments
- monocular SLAM for huge outdoor scenes using hierarchical Gaussian splats
- Real-time dense SLAM from a single camera using 3D reconstruction priors, with global pose and geometry estimation at 15 FPS
- RGB SLAM baseline using monocular depth and Gaussian scene mapping, easy to try out for 3D reconstruction from video
- visual slam experiments toolkit with easy dataset handling and benchmarking
- VGGT is a neural network that predicts 3D scene attributes like camera parameters, depth, and point maps from images
- Shows how to use OpenCV and COLMAP for simple structure from motion and 3D scene reconstruction
- procedural 3D scene generator for computer vision and AI training
- ElasticFusion lets you build real-time 3D maps of indoor spaces using an RGB-D camera, perfect for capturing detailed room layouts as you move around
- Maps Home Assistant entities to SVG floorplans
