LED2-Net: Monocular 360 layout estimation via differentiable depth rendering

Fu En Wang, Yu Hsuan Yeh, Min Sun, Wei Chen Chiu, Yi Hsuan Tsai

研究成果: Conference contribution同行評審

26 引文 斯高帕斯(Scopus)

摘要

Although significant progress has been made in room layout estimation, most methods aim to reduce the loss in the 2D pixel coordinate rather than exploiting the room structure in the 3D space. Towards reconstructing the room layout in 3D, we formulate the task of 360 layout estimation as a problem of predicting depth on the horizon line of a panorama. Specifically, we propose the Differentiable Depth Rendering procedure to make the conversion from layout to depth prediction differentiable, thus making our proposed model end-to-end trainable while leveraging the 3D geometric information, without the need of providing the ground truth depth. Our method achieves state-of-the-art performance on numerous 360 layout benchmark datasets. Moreover, our formulation enables a pre-training step on the depth dataset, which further improves the generalizability of our layout estimation model.

原文English
主出版物標題Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
發行者IEEE Computer Society
頁面12951-12960
頁數10
ISBN(電子)9781665445092
DOIs
出版狀態Published - 2021
事件2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021 - Virtual, Online, United States
持續時間: 19 6月 202125 6月 2021

出版系列

名字Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
ISSN(列印)1063-6919

Conference

Conference2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2021
國家/地區United States
城市Virtual, Online
期間19/06/2125/06/21

指紋

深入研究「LED2-Net: Monocular 360 layout estimation via differentiable depth rendering」主題。共同形成了獨特的指紋。

引用此