Surface Snapping Optimization Layer for Single Image Object Shape Reconstruction

Yuan-Ting Hu, Alex Schwing, Raymond A. Yeh
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:13599-13609, 2023.

Abstract

Reconstructing the 3D shape of objects observed in a single image is a challenging task. Recent approaches rely on visual cues extracted from a given image learned from a deep net. In this work, we leverage recent advances in monocular scene understanding to incorporate an additional geometric cue of surface normals. For this, we proposed a novel optimization layer that encourages the face normals of the reconstructed shape to be aligned with estimated surface normals. We develop a computationally efficient conjugate-gradient-based method that avoids the computation of a high-dimensional sparse matrix. We show this framework to achieve compelling shape reconstruction results on the challenging Pix3D and ShapeNet datasets.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-hu23f, title = {Surface Snapping Optimization Layer for Single Image Object Shape Reconstruction}, author = {Hu, Yuan-Ting and Schwing, Alex and Yeh, Raymond A.}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {13599--13609}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/hu23f/hu23f.pdf}, url = {https://proceedings.mlr.press/v202/hu23f.html}, abstract = {Reconstructing the 3D shape of objects observed in a single image is a challenging task. Recent approaches rely on visual cues extracted from a given image learned from a deep net. In this work, we leverage recent advances in monocular scene understanding to incorporate an additional geometric cue of surface normals. For this, we proposed a novel optimization layer that encourages the face normals of the reconstructed shape to be aligned with estimated surface normals. We develop a computationally efficient conjugate-gradient-based method that avoids the computation of a high-dimensional sparse matrix. We show this framework to achieve compelling shape reconstruction results on the challenging Pix3D and ShapeNet datasets.} }
Endnote
%0 Conference Paper %T Surface Snapping Optimization Layer for Single Image Object Shape Reconstruction %A Yuan-Ting Hu %A Alex Schwing %A Raymond A. Yeh %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-hu23f %I PMLR %P 13599--13609 %U https://proceedings.mlr.press/v202/hu23f.html %V 202 %X Reconstructing the 3D shape of objects observed in a single image is a challenging task. Recent approaches rely on visual cues extracted from a given image learned from a deep net. In this work, we leverage recent advances in monocular scene understanding to incorporate an additional geometric cue of surface normals. For this, we proposed a novel optimization layer that encourages the face normals of the reconstructed shape to be aligned with estimated surface normals. We develop a computationally efficient conjugate-gradient-based method that avoids the computation of a high-dimensional sparse matrix. We show this framework to achieve compelling shape reconstruction results on the challenging Pix3D and ShapeNet datasets.
APA
Hu, Y., Schwing, A. & Yeh, R.A.. (2023). Surface Snapping Optimization Layer for Single Image Object Shape Reconstruction. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:13599-13609 Available from https://proceedings.mlr.press/v202/hu23f.html.

Related Material