A Fast Deep Learning Network for Automatic Image Auto-Straightening
This addresses a common issue for photographers by enabling real-time, device-adapted image rotation correction across various photo categories.
The paper tackles the problem of automatically straightening images, which is challenging when horizon lines are absent, by proposing a deep learning network with rectangle-shaped depthwise convolutions and a new loss function. The method generalizes across diverse image types and significantly outperforms state-of-the-art approaches.
Rectifying the orientation of images represents a daily task for every photographer. This task may be complicated even for the human eye, especially when the horizon or other horizontal and vertical lines in the image are missing. In this paper we address this problem and propose a new deep learning network specially adapted for image rotation correction: we introduce the rectangle-shaped depthwise convolutions which are specialized in detecting long lines from the image and a new adapted loss function that addresses the problem of orientation errors. Compared to other methods that are able to detect rotation errors only on few image categories, like man-made structures, the proposed method can be used on a larger variety of photographs e.g., portraits, landscapes, sport, night photos etc. Moreover, the model is adapted to mobile devices and can be run in real time, both for pictures and for videos. An extensive evaluation of our model on different datasets shows that it remarkably generalizes, not being dependent on any particular type of image. Finally, we significantly outperform the state-of-the-art methods, providing superior results.