Said EL ABDELLAOUI1,2*, Ilham KACHBAL1
1 LAPSSII, High School of Technology, Cadi Ayyad University, B.P. 89, Safi, Morocco
Said.email@example.com (*Corresponding author)
2 LRIT Laboratory Associate Unit to CNRST (URAC 29), Faculty of Sciences,
Mohammed V-Agdal University, Rabat, Morocco
Abstract: Image matting is one of the most important tasks in the computer vision community whose popularity has increased in recent years. This is a highly critical method in video and image editing applications, which involves the separation of the foreground from the background of an image. The previous methods provide a low accuracy when the background and foreground of an image are similar. This paper proposes an effective matting method that integrates the combination of a supervised deep learning matting network generator and a self-supervised refinement network. The generator uses a supervised encoder-decoder network for the extraction of the foreground and alpha matte from the original input image. The results obtained by this network are employed by the self-supervised refinement network to evaluate the newly created composite images and ultimately improving the matting process. The proposed method has obtained better results in comparison with other methods, which makes it more reliable.
Keywords: Matting, Alpha matte, KNN, Soft-Segmentation, GAN.
>>FULL TEXT: PDF
CITE THIS PAPER AS:
Said EL ABDELLAOUI, Ilham KACHBAL, Deep Residual Network for High-Resolution Background Matting, Studies in Informatics and Control, ISSN 1220-1766, vol. 30(3), pp. 51-59, 2021. https://doi.org/10.24846/v30i3y202105