EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Multi-Head Attention Refiner for Multi-View 3D Reconstruction

Kyunghee Lee iD, Ihjoon Cho iD, Boseung Yang iD, Unsang Park iD

Journal of Imaging · 24.10.2024 · 10 (11) · 268

DOI: 10.3390/jimaging10110268 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Imaging, online erschienen am 24.10.2024. Verfasst von Kyunghee Lee und 3 weiteren Autor:innen. Beteiligte Einrichtung: Department of Computer Science and Engineering, Sogang University, 35 Baekbeom-ro (Sinsu-dong), Mapo-gu, Seoul 04107, Republic of Korea. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Korea government (MSIT).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Traditional 3D reconstruction models have consistently faced the challenge of balancing high recall of object edges with maintaining a high precision. In this paper, we introduce a post-processing method, the Multi-Head Attention Refiner (MA-R), designed to address this issue by integrating a multi-head attention mechanism into the U-Net style refiner module. Our method demonstrates improved capability in capturing intricate image details, leading to significant enhancements in boundary predictions and recall rates. In our experiments, the proposed approach notably improves the reconstruction performance of Pix2Vox++ when multiple images are used as the input. Specifically, with 20-view images, our method achieves an IoU score of 0.730, a 1.1% improvement over the 0.719 of Pix2Vox++, and a 2.1% improvement in F-Score, achieving 0.483 compared to 0.462 of Pix2Vox++. These results underscore the robustness of our approach in enhancing both precision and recall in 3D reconstruction tasks involving multiple views.

Schlagwörter

Stichwörter der Autor:innen

attention mechanismmulti-head attentionmulti-view 3D reconstructionobject boundary predictionrefiner

Autor:innen und Institutionen

  • Department of Computer Science and Engineering, Sogang University, 35 Baekbeom-ro (Sinsu-dong), Mapo-gu, Seoul 04107, Republic of Korea.

ORCID-Profile: Kyunghee Lee, Ihjoon Cho, Boseung Yang, Unsang Park

Förderung

  • Korea government (MSIT)

Zitieren

APA

Lee, K., Cho, I., Yang, B., & Park, U. (2024). Multi-Head Attention Refiner for Multi-View 3D Reconstruction. Journal of Imaging, 10(11), 268. https://doi.org/10.3390/jimaging10110268

Vancouver

Lee K, Cho I, Yang B, Park U. Multi-Head Attention Refiner for Multi-View 3D Reconstruction. Journal of Imaging. 2024;10(11):268. doi:10.3390/jimaging10110268

BibTeX
@article{Lee2024,
  author = {Lee, Kyunghee and Cho, Ihjoon and Yang, Boseung and Park, Unsang},
  title = {Multi-Head Attention Refiner for Multi-View 3D Reconstruction},
  journal = {Journal of Imaging},
  year = {2024},
  volume = {10},
  number = {11},
  pages = {268},
  doi = {10.3390/jimaging10110268}
}
Kostenlos starten →