The main objective of this approach it to extract semantically significant regions of a still image. The use of Deep Neural Networks allows the algorithm to work well also for generic images, while state-of-the-art semantic segmentation solutions are typically tailored for limited application cases. More information can be found in the paper published at the IEEE International Conference on Image Processing 2016 available from IEEE Xplore. The paper was among ICIP 2016 Best Paper / Student Paper Finalist. (For more information visit https://arxiv.org/abs/1609.09220, https://www.researchgate.net/profile/Mahdyar_Ravanbakhsh)