This project tackles the problem of vision in participating media. It aims at restoring the visual quality of images taken in environments such as fog, rain, sand and water. Initially, we are targeting the underwater medium. In other words, given an underwater image (i.e., an image acquired from an underwater scene), the objective is to extract relevant information about the objects observed in the image to estimate the transmission map. This map is a combination of the depth map and the medium attenuation coefficients. It can be applied in the image formation model to recover the visual quality of the image.



[JVCI 2018] Wagner Barros, Erickson R. Nascimento, Walysson V. Barbosa, Mario F. M. Campos. Single-Shot Underwater Image Restoration: A visual quality-aware method based on light propagation model, Journal of Visual Communication and Image Representation (JVCI), 2018.
Visit the page for more information and paper access.

[ICIP 2018] Walysson V. Barbosa, Henrique G. B. Amaral, Thiago L. Rocha, Erickson R. Nascimento. Visual-Quality-Driven Learning for Underwater Vision Enhancement, IEEE International Conference on Image Processing (ICIP), 2018. To appear!
Visit the page for the code and paper access.


[ICIP 2018] Underwater Vision Dataset. To be published!
Visit the dataset page for video info and download.


This project is supported by CNPq, CAPES, and FAPEMIG.


Walysson Vital Barbosa

MSc Student

Former Members

Thiago Lages Rocha

Undergraduate Student

Henrique Grandinetti Barbosa Amaral

Undergraduate Student

Wagner Ferreira de Barros

Professor IFMG