Fast Shadow Detection from a Single Image Using a Patched Convolutional Neural Network
This work addresses the need for efficient shadow detection in robotic systems, representing an incremental improvement in speed over existing methods.
The paper tackles the problem of slow shadow detection from single images, which is unsuitable for robotic applications, by introducing a fast method using a patched convolutional neural network combined with a shadow prior map, achieving a time complexity reduction of one to two orders of magnitude compared to state-of-the-art methods while maintaining accuracy.
In recent years, various shadow detection methods from a single image have been proposed and used in vision systems; however, most of them are not appropriate for the robotic applications due to the expensive time complexity. This paper introduces a fast shadow detection method using a deep learning framework, with a time cost that is appropriate for robotic applications. In our solution, we first obtain a shadow prior map with the help of multi-class support vector machine using statistical features. Then, we use a semantic- aware patch-level Convolutional Neural Network that efficiently trains on shadow examples by combining the original image and the shadow prior map. Experiments on benchmark datasets demonstrate the proposed method significantly decreases the time complexity of shadow detection, by one or two orders of magnitude compared with state-of-the-art methods, without losing accuracy.