Robust light field angular super-resolution via multi-dimensional feature fusion and attention-guided refinement
PLOS One · 15.07.2026 · 21 (7) · e0353845
DOI: 10.1371/journal.pone.0353845 ↗Kurz gefasst
Fachartikel in der Zeitschrift PLOS One, online erschienen am 15.07.2026. Verfasst von Xiyao Hua und 2 weiteren Autor:innen. Beteiligte Einrichtung: School of Big Data and Artificial Intelligence, Chengdu Technological University, Chengdu, Chinax. Hauptthemen laut MEDLINE: Image Processing, Computer-Assisted. Die Arbeit nennt 39 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Chengdu Technological University Doctoral Talent Program.
Abstract (Originalsprache)
Angular super-resolution (ASR) is a fundamental task in light field (LF) imaging, aimed at reconstructing a dense LF from sparsely sampled views. Despite significant progress, current methods often struggle to preserve consistency in complex scenarios such as severe occlusions and large-disparity regions. In this paper, we propose a robust attention-guided multi-dimensional feature fusion network (LFAMF) for LF angular reconstruction. The proposed framework comprises two synergistic stages: a multi-dimensional feature fusion stage and an attention-guided refinement stage. Specifically, we design a multi-stream subnetwork (MFNet) to extract intrinsic physical characteristics across the spatial, angular, EPI, and pseudo-video sequence domains. Simultaneously, a geometry-prior-based subnetwork (GSPNet) is incorporated to leverage scene structure for improved texture preservation. To effectively integrate these complementary streams, an attention-guided fusion subnetwork (AFNet) is employed to adaptively merge intermediate results. Extensive experiments on both synthetic and real-world datasets demonstrate that the LFAMF model significantly outperforms state-of-the-art methods, particularly in maintaining structural integrity at occlusion boundaries and highly textured areas while ensuring superior angular consistency.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Autor:innen und Institutionen
- School of Big Data and Artificial Intelligence, Chengdu Technological University, Chengdu, Chinax.
ORCID-Profile: Xiyao Hua
Förderung
- Chengdu Technological University Doctoral Talent Program
Zitieren
APA
Hua, X., Su, B., & Yang, D. (2026). Robust light field angular super-resolution via multi-dimensional feature fusion and attention-guided refinement. PLOS One, 21(7), e0353845. https://doi.org/10.1371/journal.pone.0353845
Vancouver
Hua X, Su B, Yang D. Robust light field angular super-resolution via multi-dimensional feature fusion and attention-guided refinement. PLOS One. 2026;21(7):e0353845. doi:10.1371/journal.pone.0353845
BibTeX
@article{Hua2026,
author = {Hua, Xiyao and Su, Boni and Yang, Daili},
title = {Robust light field angular super-resolution via multi-dimensional feature fusion and attention-guided refinement},
journal = {PLOS One},
year = {2026},
volume = {21},
number = {7},
pages = {e0353845},
doi = {10.1371/journal.pone.0353845}
}
EUVIMED