Feature-Centric Underwater Vision Models
Revista Eletrônica de Iniciação Científica em Computação · 10.07.2026 · 24 · 504-511
DOI: 10.5753/reic.2026.8511 ↗Kurz gefasst
Fachartikel in der Zeitschrift Revista Eletrônica de Iniciação Científica em Computação, online erschienen am 10.07.2026. Verfasst von Emanuel da Costa Silva und 2 weiteren Autor:innen. Die Arbeit nennt 25 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Underwater image enhancement (UIE) models are often optimized for human visual perception, which does not necessarily translate to improved performance in automated vision tasks. This paper describes a consolidated research journey consisting of the AquaFeat ecosystem: a series of plug-and-play modules designed to enhance hierarchical features for downstream robotics tasks. We present the transition from AquaFeat, a detection-focused module, to AquaFeat+, which introduces global-scale attention for classification and tracking, and finally C-Feat. C-Feat integrates these feature-centric backbones into a Compact Model framework, achieving a remarkable inference speed of 280 FPS while maintaining state-of-the-art performance in complex underwater environments.
Zitieren
APA
Silva, E. D. C., Schein, T. T., & Junior, P. L. J. D. (2026). Feature-Centric Underwater Vision Models. Revista Eletrônica de Iniciação Científica em Computação, 24, 504-511. https://doi.org/10.5753/reic.2026.8511
Vancouver
Silva EDC, Schein TT, Junior PLJD. Feature-Centric Underwater Vision Models. Revista Eletrônica de Iniciação Científica em Computação. 2026;24:504-511. doi:10.5753/reic.2026.8511
BibTeX
@article{Silva2026,
author = {Silva, Emanuel da Costa and Schein, Tatiana Taís and Junior, Paulo Lilles Jorge Drews},
title = {Feature-Centric Underwater Vision Models},
journal = {Revista Eletrônica de Iniciação Científica em Computação},
year = {2026},
volume = {24},
pages = {504--511},
doi = {10.5753/reic.2026.8511}
}
EUVIMED