TY - GEN
T1 - Towards Part-Based Understanding of RGB-D Scans
AU - Bokhovkin, Alexey
AU - Ishimtsev, Vladislav
AU - Bogomolov, Emil
AU - Zorin, Denis
AU - Artemov, Alexey
AU - Burnaev, Evgeny
AU - Dai, Angela
N1 - Funding Information:
We would like to thank the support of the Zentrum Digital-isierung.Bayern (ZD.B) and the Russian Science Foundation (Grant 19-41-04109). We additionally acknowledge the use of Skoltech CDISE HPC cluster Zhores.
Publisher Copyright:
© 2021 IEEE
PY - 2021
Y1 - 2021
N2 - Recent advances in 3D semantic scene understanding have shown impressive progress in 3D instance segmentation, enabling object-level reasoning about 3D scenes; however, a finer-grained understanding is required to enable interactions with objects and their functional understanding. Thus, we propose the task of part-based scene understanding of real-world 3D environments: from an RGB-D scan of a scene, we detect objects, and for each object predict its decomposition into geometric part masks, which composed together form the complete geometry of the observed object. We leverage an intermediary part graph representation to enable robust completion as well as building of part priors, which we use to construct the final part mask predictions. Our experiments demonstrate that guiding part understanding through part graph to part prior-based predictions significantly outperforms alternative approaches to the task of semantic part completion.
AB - Recent advances in 3D semantic scene understanding have shown impressive progress in 3D instance segmentation, enabling object-level reasoning about 3D scenes; however, a finer-grained understanding is required to enable interactions with objects and their functional understanding. Thus, we propose the task of part-based scene understanding of real-world 3D environments: from an RGB-D scan of a scene, we detect objects, and for each object predict its decomposition into geometric part masks, which composed together form the complete geometry of the observed object. We leverage an intermediary part graph representation to enable robust completion as well as building of part priors, which we use to construct the final part mask predictions. Our experiments demonstrate that guiding part understanding through part graph to part prior-based predictions significantly outperforms alternative approaches to the task of semantic part completion.
UR - http://www.scopus.com/inward/record.url?scp=85123214705&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85123214705&partnerID=8YFLogxK
U2 - 10.1109/CVPR46437.2021.00740
DO - 10.1109/CVPR46437.2021.00740
M3 - Conference contribution
AN - SCOPUS:85123214705
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 7480
EP - 7490
BT - Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
PB - IEEE Computer Society
T2 - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
Y2 - 19 June 2021 through 25 June 2021
ER -