The underwater world has gained significant attention in research in recent years, particularly in the context of ocean exploration. Images serve as a valuable data source for underwater tasks, but they face several issues related to light behavior in this environment. Given the complexity of capturing data from the sea and the large variability of environmental components (depth, distance, suspended particles, turbidity, etc.), synthesized underwater scenes can provide relevant data to improve image processing algorithms and computer vision tasks. The main goal of this survey is to summarize techniques to underwater image synthesis, their contributions and correlations, and to highlight further directions and opportunities in this research domain.
1 December 2020 in Rio Oil and Gas Expo and Conference
Modelos de Aprendizado de Máquina, subárea da Inteligência Artificial, vêm se tornando cada vez mais comuns no cotidiano. Contudo, há a necessidade de analisar o impacto que esses modelos podem ter em minorias socialmente vulneráveis e os vieses que podem ser apresentados, independente de sua causa e mitigando sua consequência. Para isso, é importante entender incidentes que já ocorreram, observando suas potenciais falhas técnicas e consequências jurídicas relacionadas. A partir dessa análise, o trabalho propõe soluções que podem minimizar o impacto das falhas.
About Me
Researcher on Computer Vision and Artificial Intelligence.
Researcher on Computer Vision and Artificial Intelligence.