Skip to content
DraconiticResearch
Search Sources API
190,860 studies31 sourcesUpdated 37 min ago
Peer-reviewedCC BY 4.0Mathematics

Cross-Domain Mesh–Progressive Fusion for Light Field Reconstruction

  • Ebrahem Elkady1,2
  • Ahmed Salem3
  • Hyun-Soo Kang4
  • Jae-Won Suh1
4 affiliations
  1. 1School of Electronics Engineering, College of Electrical and Computer Engineering, Chungbuk National University, Cheongju 28644, Republic of Korea
  2. 2Information Technology Department, Faculty of Computers and Information, Assiut University, Assiut 71526, Egypt
  3. 3Electrical Engineering Department, Faculty of Engineering, Assiut University, Assiut 71526, Egypt
  4. 4School of Information and Communication Engineering, College of Electrical and Computer Engineering, Chungbuk National University, Cheongju 28644, Republic of Korea
Open at sourceDOICite

Abstract

Light field angular super-resolution (LFASR) seeks to recover missing viewpoints from sparsely captured angular samples, while preserving the underlying scene structure across views. This task is complicated by disparity-induced variations, incomplete visibility near occlusion boundaries, and view-dependent appearance changes caused by reflective surfaces. In this work, we propose a cross-domain EPI–spatial–angular (X-ESA) framework that combines EPI-based geometric modeling with complementary spatial and angular feature extraction. A CSWin-based EPI transformer captures long-range, slope-aligned dependencies, while dedicated convolutional branches refine spatial details and cross-view relationships. These features are integrated through a mesh–progressive fusion strategy, followed by channel attention and spatial residual refinement. Experiments on real-world and synthetic datasets under the 2×2→7×7 setting result in average performances of 41.29 dB/0.989 SSIM and 40.98 dB/0.989 SSIM, respectively. The proposed model contains 2.84 M trainable parameters and provides competitive reconstruction quality relative to existing LFASR methods.

Related

  1. Strict unimodality of the exact time-domain diffuse-light response under a Robin boundary condition
    Junyong Eom, Jinwan Park
    arXivPreprintCC BY-NC-ND 4.0Full textAnalysis & PDEsMathematical physicsarXiv
  2. Lifetime-unified quadratic and cubic approximations of fluorescence peak time for a time-domain diffuse-light response
    Junyong Eom
    arXivPreprintarXiv licenceFull textAnalysis & PDEsarXiv
  3. Progressive Reliability-Guided Refinement with Cross-Modal Consistency and Iterative Correction for Weakly Supervised PET/CT Tumor Segmentation
    Zhaoshuo Diao, Huiyan Jiang
    MathematicsPeer-reviewedCC BY 4.0General mathematicsCrossref
  4. IV-PAFNet: Infrared-visible cross-modal alignment and fusion network for nighttime UAV object detection
    Cheng Yuan, Shichang Shang, Jinghao Zhang +2
    PLOS OnePeer-reviewedCC BYFull textBioinformatics & computational biologyNeurosciencePLOSEurope PMCPMC
  5. Structure-Driven Methodology: An Emerging Cross-Domain Paradigm
    Shengchang Chen
    arXivPreprintCC BY-NC-SA 4.0Full textGeophysics & seismologyGeneral mathematicsarXiv

Cite

BibTeXRIS
BibTeX
@article{elkady2026crossdomain,
  title = {{Cross-Domain Mesh–Progressive Fusion for Light Field Reconstruction}},
  author = {Ebrahem Elkady and Ahmed Salem and Hyun-Soo Kang and Jae-Won Suh},
  journal = {Mathematics},
  year = {2026},
  month = oct,
  doi = {10.3390/math14203811},
  url = {https://www.mdpi.com/2227-7390/14/20/3811},
  note = {CC BY 4.0},
}
RIS
TY  - JOUR
TI  - Cross-Domain Mesh–Progressive Fusion for Light Field Reconstruction
AU  - Ebrahem Elkady
AU  - Ahmed Salem
AU  - Hyun-Soo Kang
AU  - Jae-Won Suh
PY  - 2026
DA  - 2026/10/20
JO  - Mathematics
DO  - 10.3390/math14203811
UR  - https://www.mdpi.com/2227-7390/14/20/3811
AB  - Light field angular super-resolution (LFASR) seeks to recover missing viewpoints from sparsely captured angular samples, while preserving the underlying scene structure across views. This task is complicated by disparity-induced variations, incomplete visibility near occlusion boundaries, and view-dependent appearance changes caused by reflective surfaces. In this work, we propose a cross-domain EPI–spatial–angular (X-ESA) framework that combines EPI-based geometric modeling with complementary spatial and angular feature extraction. A CSWin-based EPI transformer captures long-range, slope-aligned dependencies, while dedicated convolutional branches refine spatial details and cross-view relationships. These features are integrated through a mesh–progressive fusion strategy, followed by channel attention and spatial residual refinement. Experiments on real-world and synthetic datasets under the 2×2→7×7 setting result in average performances of 41.29 dB/0.989 SSIM and 40.98 dB/0.989 SSIM, respectively. The proposed model contains 2.84 M trainable parameters and provides competitive reconstruction quality relative to existing LFASR methods.
SN  - 2227-7390
ER  -