EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Distortion-Aware Bi-Projection Fusion for 360∘ Monocular Depth Estimation via Coordinate Attention

Lichuan Geng, Li Ma, Yongzhi Qin, Chenyang He, Peng Sun

Electronics · 13.07.2026 · 15 (14) · 3066

DOI: 10.3390/electronics15143066 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 13.07.2026. Verfasst von Lichuan Geng und 4 weiteren Autor:innen. Die Arbeit nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Monocular depth estimation for 360∘ omnidirectional images is essential for immersive scene understanding and 3D reconstruction but remains challenging due to the non-uniform geometric distortions introduced by equirectangular projection (ERP). In particular, ERP suffers from latitude-dependent sampling bias, where the effective receptive field of standard convolution varies with the spherical latitude, leading to inconsistent feature representation and degraded depth prediction in high-distortion regions. To address this problem, this paper proposes a distortion-aware bi-projection fusion framework that integrates the global contextual continuity of ERP with the locally distortion-reduced geometric representation of cube map projection (CMP). The core component of the proposed framework is a Multi-scale Coordinate-Transformer Fusion (MCTF) module, which combines convolutional feature mixing, Transformer-based global context modeling, and Coordinate Attention-based spatial recalibration. By explicitly encoding vertical coordinate information into the fusion process, MCTF adaptively recalibrates feature responses according to latitude-dependent distortion patterns. Extensive experiments on the 3D60, Matterport3D, and Stanford2D3D benchmarks demonstrate that the proposed method consistently outperforms state-of-the-art omnidirectional depth estimation approaches. On the 3D60 dataset, our method reduces RMSE by 21.2% compared with the ERP baseline and achieves a 30% RMSE reduction in polar regions, where ERP distortion is most severe. These results validate the effectiveness of coordinate-aware feature calibration for robust 360∘ monocular depth estimation.

Zitieren

APA

Geng, L., Ma, L., Qin, Y., He, C., & Sun, P. (2026). Distortion-Aware Bi-Projection Fusion for 360∘ Monocular Depth Estimation via Coordinate Attention. Electronics, 15(14), 3066. https://doi.org/10.3390/electronics15143066

Vancouver

Geng L, Ma L, Qin Y, He C, Sun P. Distortion-Aware Bi-Projection Fusion for 360∘ Monocular Depth Estimation via Coordinate Attention. Electronics. 2026;15(14):3066. doi:10.3390/electronics15143066

BibTeX
@article{Geng2026,
  author = {Geng, Lichuan and Ma, Li and Qin, Yongzhi and He, Chenyang and Sun, Peng},
  title = {Distortion-Aware Bi-Projection Fusion for 360∘ Monocular Depth Estimation via Coordinate Attention},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {14},
  pages = {3066},
  doi = {10.3390/electronics15143066}
}
Kostenlos starten →