Vanishing Point Detection with Direct and Transposed Fast Hough Transform inside the neural network
This addresses the problem of accurate vanishing point detection for applications like computer vision and autonomous driving, but it is incremental as it builds on existing Hough transform methods.
The paper tackles vanishing point detection in images by proposing a new neural network architecture that integrates direct and transposed Fast Hough Transforms, enabling direct output in input image coordinates for simple maximum selection. It demonstrates superiority over existing methods on a DVR image dataset.
In this paper, we suggest a new neural network architecture for vanishing point detection in images. The key element is the use of the direct and transposed Fast Hough Transforms separated by convolutional layer blocks with standard activation functions. It allows us to get the answer in the coordinates of the input image at the output of the network and thus to calculate the coordinates of the vanishing point by simply selecting the maximum. Besides, it was proved that calculation of the transposed Fast Hough Transform can be performed using the direct one. The use of integral operators enables the neural network to rely on global rectilinear features in the image, and so it is ideal for detecting vanishing points. To demonstrate the effectiveness of the proposed architecture, we use a set of images from a DVR and show its superiority over existing methods. Note, in addition, that the proposed neural network architecture essentially repeats the process of direct and back projection used, for example, in computed tomography.