%0 Conference Paper %1 Ulusoy:2015:TPV:2848895.2848907 %A Ulusoy, Ali Osman %A Geiger, Andreas %A Black, Michael J. %B Proceedings of the 2015 International Conference on 3D Vision %C Washington, DC, USA %D 2015 %I IEEE Computer Society %K 2015 3D computer-vision germany probability reconstruction %P 10--18 %R 10.1109/3DV.2015.9 %T Towards Probabilistic Volumetric Reconstruction Using Ray Potentials %U https://dl.acm.org/citation.cfm?id=2848907 %X This paper presents a novel probabilistic foundation for volumetric 3D reconstruction. We formulate the problem as inference in a Markov random field, which accurately captures the dependencies between the occupancy and appearance of each voxel, given all input images. Our main contribution is an approximate highly parallelized discrete-continuous inference algorithm to compute the marginal distributions of each voxel's occupancy and appearance. In contrast to the MAP solution, marginals encode the underlying uncertainty and ambiguity in the reconstruction. Moreover, the proposed algorithm allows for a Bayes optimal prediction with respect to a natural reconstruction loss. We compare our method to two state-of-the-art volumetric reconstruction algorithms on three challenging aerial datasets with LIDAR ground truth. Our experiments demonstrate that the proposed algorithm compares favorably in terms of reconstruction accuracy and the ability to expose reconstruction uncertainty. %@ 978-1-4673-8332-5

@inproceedings{Ulusoy:2015:TPV:2848895.2848907, abstract = {This paper presents a novel probabilistic foundation for volumetric 3D reconstruction. We formulate the problem as inference in a Markov random field, which accurately captures the dependencies between the occupancy and appearance of each voxel, given all input images. Our main contribution is an approximate highly parallelized discrete-continuous inference algorithm to compute the marginal distributions of each voxel's occupancy and appearance. In contrast to the MAP solution, marginals encode the underlying uncertainty and ambiguity in the reconstruction. Moreover, the proposed algorithm allows for a Bayes optimal prediction with respect to a natural reconstruction loss. We compare our method to two state-of-the-art volumetric reconstruction algorithms on three challenging aerial datasets with LIDAR ground truth. Our experiments demonstrate that the proposed algorithm compares favorably in terms of reconstruction accuracy and the ability to expose reconstruction uncertainty.}, acmid = {2848907}, added-at = {2019-03-16T16:57:55.000+0100}, address = {Washington, DC, USA}, author = {Ulusoy, Ali Osman and Geiger, Andreas and Black, Michael J.}, biburl = {https://www.bibsonomy.org/bibtex/21dbf23ec79109642f2ee6cfcaaf9d926/analyst}, booktitle = {Proceedings of the 2015 International Conference on 3D Vision}, description = {Towards Probabilistic Volumetric Reconstruction Using Ray Potentials}, doi = {10.1109/3DV.2015.9}, interhash = {3acccc4d3e45e658d4407748ec4df6ed}, intrahash = {1dbf23ec79109642f2ee6cfcaaf9d926}, isbn = {978-1-4673-8332-5}, keywords = {2015 3D computer-vision germany probability reconstruction}, numpages = {9}, pages = {10--18}, publisher = {IEEE Computer Society}, series = {3DV '15}, timestamp = {2019-03-16T16:57:55.000+0100}, title = {Towards Probabilistic Volumetric Reconstruction Using Ray Potentials}, url = {https://dl.acm.org/citation.cfm?id=2848907}, year = 2015 }