Endo-Depth-and-Motion: Reconstruction and Tracking in Endoscopic Videos скачать в хорошем качестве

Endo-Depth-and-Motion: Reconstruction and Tracking in Endoscopic Videos

Endo-Depth-and-Motion: Reconstruction and Tracking in Endoscopic Videos using Depth Networks and Photometric Constraints Authors: David Recasens, José Lamarca, José M. Fácil, J. M. M. Montiel and Javier Civera | University of Zaragoza RA-L journal paper: https://ieeexplore.ieee.org/abstract/... arXiv paper: https://arxiv.org/abs/2103.16525 Paper project page: https://davidrecasens.github.io/EndoD... Endo-Depth-and-Motion is a pipeline that estimates the 6-degrees-of-freedom camera pose anddense 3D scene models from monocular endoscopic sequences.Our approach leverages recent advances in self-superviseddepth networks to generatepseudo-RGBD frames, then tracks the camera pose using photometric residuals and fuses theregistered depth maps in a volumetric representation. Hamlyn videos: 0:00 Dataset 1 1:10 Dataset 15 1:20 Dataset 17 1:40 Dataset 20 2:03 Dataset 21 2:28 Dataset 22 3:47 Dataset 25 4:57 Dataset 26 5:24 Dataset 27 This work was supported by EndoMapper GA 863146 (EU-H2020), PGC2018-096367-B-I00 (Spanish Government), DGA-T45 17R/FSE (Aragón Government).