EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Learnable differencing center for nighttime depth perception

Zhiqiang Yan, Yupeng Zheng, Deng-Ping Fan, Xiang Li, Jun Li, Jian Yang

Visual Intelligence · 19.06.2024 · 2 (1)

DOI: 10.1007/s44267-024-00048-9 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Visual Intelligence, online erschienen am 19.06.2024. Verfasst von Zhiqiang Yan und 5 weiteren Autor:innen. Die Arbeit wurde bisher 15-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Young Scientists Fund of the National Natural Science Foundation of China, Fundamental Research Funds for the Central Universities, Postgraduate Research and Practice Innovation Program of Jiangsu Province.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Depth completion is the task of recovering dense depth map from sparse ones, usually with the help of color images. Existing image guided methods perform well on daytime depth perception self-driving benchmarks, but struggle in nighttime scenarios with poor visibility and complex illumination. To address these challenges, we propose a simple yet effective learnable differencing center network (LDCNet). The key idea is to use recurrent inter-convolution differencing (RICD) and illumination affinitive intra-convolution differencing (IAICD) to enhance the nighttime color images and reduce the negative effects of the varying illumination, respectively. RICD explicitly estimates global illumination by differencing two convolutions with different kernels, treating the small-kernel-convolution feature as the center of the large-kernel-convolution feature in a new perspective. IAICD softly alleviates the local relative light intensity by differencing a single convolution, where the center is dynamically aggregated based on neighboring pixels and the estimated illumination map in the RICD. On both nighttime depth completion and depth estimation tasks, extensive experiments demonstrate the effectiveness of our LDCNet, reaching the state of the art.

Förderung

  • Young Scientists Fund of the National Natural Science Foundation of China
  • Fundamental Research Funds for the Central Universities
  • Postgraduate Research and Practice Innovation Program of Jiangsu Province
  • National Science Fund of China

Zitieren

APA

Yan, Z., Zheng, Y., Fan, D., Li, X., Li, J., & Yang, J. (2024). Learnable differencing center for nighttime depth perception. Visual Intelligence, 2(1). https://doi.org/10.1007/s44267-024-00048-9

Vancouver

Yan Z, Zheng Y, Fan D, Li X, Li J, Yang J. Learnable differencing center for nighttime depth perception. Visual Intelligence. 2024;2(1). doi:10.1007/s44267-024-00048-9

BibTeX
@article{Yan2024,
  author = {Yan, Zhiqiang and Zheng, Yupeng and Fan, Deng-Ping and Li, Xiang and Li, Jun and Yang, Jian},
  title = {Learnable differencing center for nighttime depth perception},
  journal = {Visual Intelligence},
  year = {2024},
  volume = {2},
  number = {1},
  doi = {10.1007/s44267-024-00048-9}
}
Kostenlos starten →