It supports TSDF with voxel hashing and incremental Gaussian Splatting. Integration of depth prediction models within the SLAM pipeline. These include DepthPro, DepthAnythingV2, RAFT-Stereo, CREStereo, etc. Add
KinectFusion [27] was the first algorithm to deploy real-time volumetric truncated signed distance fields (TSDF). KinectFusion played a pivotal role in the later advancements of the GPU-accelerated TSDF family of algorithms for scene reconstruction using depth images, leading to works by [28,29]...