Register now After registration you will be able to apply for this opportunity online.
Warping Voxel-grids for 3D Robotic Mapping
This project aims to integrate loop closure optimization into voxel-based 3D mapping by developing a method to warp the voxel grid in response to updated camera trajectories. This approach eliminates the need to rebuild the map from scratch, enhancing the efficiency and adaptability of 3D mapping in real-world scenarios.
Keywords: 3D mapping, robot, voxel grid
The goal of this project is to incorporate a loop closure optimization process into voxel-based 3D mapping approaches [1]. These approaches are widely used for reconstructing the environments of robotic agents due to their incremental and fast nature. However, most existing methods rely on pre-estimating the camera trajectory, complicating their application in real-world scenarios. While SLAM-based camera pose estimation enables simultaneous mapping, loop closure optimization - where the entire trajectory may be adjusted - typically requires rebuilding the voxel map from scratch. This project aims to develop a method for "warping" the voxel grid to align with optimized camera trajectories, eliminating the need for complete reintegration and improving the efficiency and adaptability of voxel-based 3D mapping systems.
[1] Miao, Y., Armeni, I., Pollefeys, M. and Barath, D. Volumetric semantically consistent 3d panoptic mapping. IROS 2024.
The goal of this project is to incorporate a loop closure optimization process into voxel-based 3D mapping approaches [1]. These approaches are widely used for reconstructing the environments of robotic agents due to their incremental and fast nature. However, most existing methods rely on pre-estimating the camera trajectory, complicating their application in real-world scenarios. While SLAM-based camera pose estimation enables simultaneous mapping, loop closure optimization - where the entire trajectory may be adjusted - typically requires rebuilding the voxel map from scratch. This project aims to develop a method for "warping" the voxel grid to align with optimized camera trajectories, eliminating the need for complete reintegration and improving the efficiency and adaptability of voxel-based 3D mapping systems.
[1] Miao, Y., Armeni, I., Pollefeys, M. and Barath, D. Volumetric semantically consistent 3d panoptic mapping. IROS 2024.
