Automatic Photo Orientation Detection with Convolutional Neural Networks
This addresses a practical issue for digitizing analog photos, but the method is incremental as it applies existing CNNs to this specific task.
The paper tackled the problem of automatically detecting the correct orientation (0, 90, 180, or 270 degrees) of consumer photos, especially for digitizing analog photographs, and achieved substantial improvement over the state-of-the-art on a standard dataset.
We apply convolutional neural networks (CNN) to the problem of image orientation detection in the context of determining the correct orientation (from 0, 90, 180, and 270 degrees) of a consumer photo. The problem is especially important for digitazing analog photographs. We substantially improve on the published state of the art in terms of the performance on one of the standard datasets, and test our system on a more difficult large dataset of consumer photos. We use Guided Backpropagation to obtain insights into how our CNN detects photo orientation, and to explain its mistakes.