摘要
Incorporating polarization in computer vision tasks provides new solutions to high-level analytics,in particular when coupled with machine learning frameworks such as convolutional neural networks(CNN).A recent review in Opto-Electronic Science reports on the developments in data-driven polarimetric imaging,including polarimetric descattering,3D imaging,reflection removal,target detection and biomedical imaging.The review carefully analyzes these new trends with their advantages and disadvantages,and provides a general insight for future research and development.
