Sammanfattning
The present Multi-view stereo (MVS) methods with supervised learning-based networks have an impressive performance comparing with traditional MVS methods. However, the ground-truth depth maps for training are hard to be obtained and are within limited kinds of scenarios. In this paper, we propose a novel unsupervised multi-metric MVS network, named M 3 VSNet, for dense point cloud reconstruction without any supervision. To improve the robustness and completeness of point cloud reconstruction, we propose a novel multi-metric loss function that combines pixel-wise and feature-wise loss function to learn the inherent constraints from different perspectives of matching correspondences. Besides, we also incorporate the normal-depth consistency in the 3D point cloud format to improve the accuracy and continuity of the estimated depth maps. Experimental results show that M 3 VSNet establishes the state-of-the-arts unsupervised method and achieves better performance than previous supervised MVSNet on the DTU dataset and demonstrates the powerful generalization ability on the Tanks & Temples benchmark with effective improvement.
Originalspråk | engelska |
---|---|
Titel på värdpublikation | Proceeding 2021 IEEE International Conference on Image Processing (ICIP) |
Förlag | IEEE - Institute of Electrical and Electronics Engineers Inc. |
Sidor | 3163-3167 |
ISBN (elektroniskt) | 978-1-6654-4115-5 |
DOI | |
Status | Published - 2021 sep. 19 |
Externt publicerad | Ja |
Evenemang | 2021 IEEE International Conference on Image Processing (ICIP) - Anchorage, AK, USA Varaktighet: 2021 sep. 19 → 2021 sep. 22 |
Konferens
Konferens | 2021 IEEE International Conference on Image Processing (ICIP) |
---|---|
Period | 2021/09/19 → 2021/09/22 |
Ämnesklassifikation (UKÄ)
- Datorseende och robotik (autonoma system)