IPDCN2: Improvised Patch-based Deep CNN for facial retouching detection

作者：

Highlights：

• We propose an Improvised Patch-based Deep CNN for facial retouching detection.

• The input image is processed to extract 6 patches using 68 facial landmarks.

• Our model uses dual skip connection to maximize the information flow.

• The proposed scheme outperforms the existing schemes on ND-IIITD dataset.

摘要

•We propose an Improvised Patch-based Deep CNN for facial retouching detection.•The input image is processed to extract 6 patches using 68 facial landmarks.•Our model uses dual skip connection to maximize the information flow.•The proposed scheme outperforms the existing schemes on ND-IIITD dataset.

论文关键词：Multimedia forensics,Facial retouching,Convolution neural network,Residual skip connection

论文评审过程：Received 30 May 2021, Revised 15 August 2022, Accepted 15 August 2022, Available online 17 August 2022, Version of Record 23 August 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.118612