EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Adapting Convolutions for Effective Omnidirectional Image Processing

Romulo Marconato Stringhini, Thiago Lopes Trugillo Da Silveira, Claudio Rosito Jung

Journal of the Brazilian Computer Society · 29.05.2026 · 32 (1) · 1471-1485

DOI: 10.5753/jbcs.2026.5654 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of the Brazilian Computer Society, online erschienen am 29.05.2026. Verfasst von Romulo Marconato Stringhini und 2 weiteren Autor:innen. Die Arbeit nennt 116 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Omnidirectional images present unique challenges for traditional convolutional neural networks (CNNs) due to the non-uniform sampling inherent in the equirectangular projection (ERP). This projection introduces distortions, especially near the poles, and conventional fixed-size kernels in planar CNNs are not designed to handle these distortions effectively. To address this issue, we previously introduced the Spherically-Weighted Horizontally Dilated Convolutions (SWHDC) module which adjusts dilated convolutions by applying appropriate weights to account for the varying sampling density across ERP image rows. In this extended work, we provide a more comprehensive evaluation of the SWHDC module by benchmarking it against several state-of-the-art methods on the 3D object classification task. Additionally, we integrate SWHDCs into different backbones to further investigate their effectiveness in tackling the gravity alignment problem. Experimental results confirm that our approach not only improves classification accuracy but also enhances gravity alignment performances without increasing the amount of trainable parameters of the baseline backbones. These findings further establish SWHDC as a robust alternative for processing omnidirectional images in different visual computing applications.

Zitieren

APA

Stringhini, R. M., Silveira, T. L. T. D., & Jung, C. R. (2026). Adapting Convolutions for Effective Omnidirectional Image Processing. Journal of the Brazilian Computer Society, 32(1), 1471-1485. https://doi.org/10.5753/jbcs.2026.5654

Vancouver

Stringhini RM, Silveira TLTD, Jung CR. Adapting Convolutions for Effective Omnidirectional Image Processing. Journal of the Brazilian Computer Society. 2026;32(1):1471-1485. doi:10.5753/jbcs.2026.5654

BibTeX
@article{Stringhini2026,
  author = {Stringhini, Romulo Marconato and Silveira, Thiago Lopes Trugillo Da and Jung, Claudio Rosito},
  title = {Adapting Convolutions for Effective Omnidirectional Image Processing},
  journal = {Journal of the Brazilian Computer Society},
  year = {2026},
  volume = {32},
  number = {1},
  pages = {1471--1485},
  doi = {10.5753/jbcs.2026.5654}
}
Kostenlos starten →