Video Upright Adjustment and Stabilization

Abstract

We propose a novel video upright adjustment method that can reliably correct slanted video contents. Our approach combines deep learning and Bayesian inference to estimate accurate rotation angles from video frames. We train a convolutional neural network to obtain initial estimates of the rotation angles of input video frames. The initial estimates are temporally inconsistent and inaccurate. To resolve this, we use Bayesian inference. We analyze estimation errors of the network, and derive an error model. Based on the error model, we formulate video upright adjustment as a maximum a posteriori problem where we estimate consistent rotation angles from the initial estimates. Finally, we propose a joint approach to video stabilization and upright adjustment to minimize information loss. Experimental results show that our video upright adjustment method can effectively correct slanted video contents, and our joint approach can achieve visually pleasing results from shaky and slanted videos. © 2019. The copyright of this document resides with its authors. It may be distributed unchanged freely in print or electronic forms.