TY - GEN
T1 - GTT-Net
T2 - 18th IEEE/CVF International Conference on Computer Vision, ICCV 2021
AU - Xu, Xiangyu
AU - Dunn, Enrique
N1 - Publisher Copyright:
© 2021 IEEE
PY - 2021
Y1 - 2021
N2 - We present GTT-Net, a supervised learning framework for the reconstruction of sparse dynamic 3D geometry. We build on a graph-theoretic formulation of the generalized trajectory triangulation problem, where non-concurrent multi-view imaging geometry is known but global image sequencing is not provided. GTT-Net learns pairwise affinities modeling the spatio-temporal relationships among our input observations and leverages them to determine 3D geometry estimates. Experiments reconstructing 3D motion-capture sequences show GTT-Net outperforms the state of the art in terms of accuracy and robustness. Within the context of articulated motion reconstruction, our proposed architecture is 1) able to learn and enforce semantic 3D motion priors for shared training and test domains, while being 2) able to generalize its performance across different training and test domains. Moreover, GTT-Net provides a computationally streamlined framework for trajectory triangulation with applications to multi-instance reconstruction and event segmentation.
AB - We present GTT-Net, a supervised learning framework for the reconstruction of sparse dynamic 3D geometry. We build on a graph-theoretic formulation of the generalized trajectory triangulation problem, where non-concurrent multi-view imaging geometry is known but global image sequencing is not provided. GTT-Net learns pairwise affinities modeling the spatio-temporal relationships among our input observations and leverages them to determine 3D geometry estimates. Experiments reconstructing 3D motion-capture sequences show GTT-Net outperforms the state of the art in terms of accuracy and robustness. Within the context of articulated motion reconstruction, our proposed architecture is 1) able to learn and enforce semantic 3D motion priors for shared training and test domains, while being 2) able to generalize its performance across different training and test domains. Moreover, GTT-Net provides a computationally streamlined framework for trajectory triangulation with applications to multi-instance reconstruction and event segmentation.
UR - http://www.scopus.com/inward/record.url?scp=85127766910&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85127766910&partnerID=8YFLogxK
U2 - 10.1109/ICCV48922.2021.00574
DO - 10.1109/ICCV48922.2021.00574
M3 - Conference contribution
AN - SCOPUS:85127766910
T3 - Proceedings of the IEEE International Conference on Computer Vision
SP - 5775
EP - 5784
BT - Proceedings - 2021 IEEE/CVF International Conference on Computer Vision, ICCV 2021
Y2 - 11 October 2021 through 17 October 2021
ER -