Comparing visual descriptors and automatic rating strategies for video aesthetics prediction

作者：

Highlights：

• A procedure for automatically predicting video aesthetics perception is proposed.

• We annotate our corpus in an automatic way by means of YouTube metadata.

• Quality-based and quantity-based annotations are evaluated and compared.

• 8 families of descriptors are compared to identify the most valuable ones.

• 62.3% accuracy with only 2 descriptors and 64.9% combining the 4 best families.

摘要

Highlights•A procedure for automatically predicting video aesthetics perception is proposed.•We annotate our corpus in an automatic way by means of YouTube metadata.•Quality-based and quantity-based annotations are evaluated and compared.•8 families of descriptors are compared to identify the most valuable ones.•62.3% accuracy with only 2 descriptors and 64.9% combining the 4 best families.

论文关键词：Automatic aesthetics prediction,Image descriptors,Video descriptors,YouTube,Automatic annotation

论文评审过程：Received 1 February 2015, Revised 11 June 2016, Accepted 13 July 2016, Available online 15 July 2016, Version of Record 27 July 2016.

论文官网地址：https://doi.org/10.1016/j.image.2016.07.004