Vanishing point detection with direct and transposed fast Hough transform inside the neural network

Abstract

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.

On a Fast Hough/Radon Transform as a Compact Summation Scheme over Digital Straight Line Segments

“…Such networks include an untrainable layer that calculates the discretization of the Radon transform. This layer allows for the extraction of information about lines and segments in an image, even if the receptive fields of convolutional neurons are small, unlike classical convolutional layers [28][29][30]. This effect is most easily demonstrated by comparing neural network line and segment search methods [31,32].…”