EUVIMEDEUVIMED entdecken
Crossref

Reinforcing Local Structure Perception for Monocular Depth Estimation

Siyuan Zhang, Jingxian Dong, Aliaksandr Chervan, Dzmitry Kurlovich, Wenguang Hou, Mingyue Ding

IEEE Sensors Journal · 15.08.2023 · 23 (16) · 18539-18549

DOI: 10.1109/jsen.2023.3293156 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Sensors Journal, erschienen am 15.08.2023. Verfasst von Siyuan Zhang und 5 weiteren Autor:innen. Die Arbeit wurde bisher 7-mal zitiert (Crossref) und nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monocular depth estimation is a basic and critical task in computer vision that finds wide applications in various domains, including robot navigation and autonomous driving. A prevailing method nowadays is leveraging hybrid depth datasets obtained from various depth sensors to predict affine-invariant depth under supervised learning. However, the varying depth ranges in hybrid datasets can result in an unstable network. While some affine-invariant loss functions have been introduced, existing methods may lead to suboptimal geometric structures, such as blurred boundaries and details. To tackle this issue, our approach is centered on reinforcing the local structural perception of images. Specifically, we propose a novel pixel-level supervised loss, called the windowed correlation regression (WCR) loss. It computes the windowed Pearson correlation coefficient (PCC) to constrain the similarity of data distribution within a local region. In addition, we introduce a new coarse-to-fine multiscale normal (CFMN) loss in conjunction with the former loss to further improve geometric accuracy. Our experimental results on six zero-shot datasets demonstrate that our method outperforms state-of-the-art (SOTA) methods. In terms of local geometric structural precision, our method achieves sharper edges and more consistent local grayscale.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Zhang, S., Dong, J., Chervan, A., Kurlovich, D., Hou, W., & Ding, M. (2023). Reinforcing Local Structure Perception for Monocular Depth Estimation. IEEE Sensors Journal, 23(16), 18539-18549. https://doi.org/10.1109/jsen.2023.3293156

Vancouver

Zhang S, Dong J, Chervan A, Kurlovich D, Hou W, Ding M. Reinforcing Local Structure Perception for Monocular Depth Estimation. IEEE Sensors Journal. 2023;23(16):18539-18549. doi:10.1109/jsen.2023.3293156

BibTeX
@article{Zhang2023,
  author = {Zhang, Siyuan and Dong, Jingxian and Chervan, Aliaksandr and Kurlovich, Dzmitry and Hou, Wenguang and Ding, Mingyue},
  title = {Reinforcing Local Structure Perception for Monocular Depth Estimation},
  journal = {IEEE Sensors Journal},
  year = {2023},
  volume = {23},
  number = {16},
  pages = {18539--18549},
  doi = {10.1109/jsen.2023.3293156}
}
Kostenlos starten →