Learning structure of stereoscopic image for no-reference quality assessment with convolutional neural network

作者：

Highlights：

• CNNs are employed to learn the local structures for stereoscopic image quality assessment.

• Two CNNs are designed to learn the image local structures based on different inputs.

• CNN parameters are pretrained on 2D images and transferred to stereoscopic images.

• The performances on public databases demonstrate the superiority of proposed model.

摘要

Highlights•CNNs are employed to learn the local structures for stereoscopic image quality assessment.•Two CNNs are designed to learn the image local structures based on different inputs.•CNN parameters are pretrained on 2D images and transferred to stereoscopic images.•The performances on public databases demonstrate the superiority of proposed model.

论文关键词：Stereoscopic image,Quality assessment,Convolutional neural network (CNN)

论文评审过程：Received 4 August 2015, Revised 24 January 2016, Accepted 31 January 2016, Available online 6 February 2016, Version of Record 23 August 2016.

论文官网地址：https://doi.org/10.1016/j.patcog.2016.01.034