Assembling Convolution Neural Networks for Automatic Viewing Transformation

Abstract

Images taken under different camera poses are rotated or distorted, which leads to poor perception experiences. This paper proposes a new framework to automatically transform the images to the conformable view setting by assembling different convolution neural networks. Specifically, a referential 3D ground plane is firstly derived from the RGB image and a novel projection mapping algorithm is developed to achieve automatic viewing transformation. Extensive experimental results demonstrate that the proposed method outperforms the state-ofthe-art vanishing points based methods by a large margin in terms of accuracy and robustness.

Blind First-Order Perspective Distortion Correction Using Parallel Convolutional Neural Networks

“…Blind distortion correction is an ill-posed problem. Therefore, learning-based methods using only a single distorted image are being pursued [ 10 , 11 , 12 , 42 , 43 , 44 , 45 , 46 ]. Deep learning for correcting documents were proposed recently [ 12 , 44 , 45 , 46 ] which implements convolutional neural networks, encoder-decoders, and U-net-based architectures [ 47 ].…”

A neural refinement network for single image view synthesis

“…Depth is one of the most common projection types since there exists an explicit geometric relationship between the image coordinates in the source and target images for projection source view target view IBR result from [1] our result using depth. Most previous methods [2,[12][13][14][15] predict the depth of the input image to forward warp pixels of the source image. However, several source pixels might be mapped to one target location, which in turn will result in holes for pixels at unassigned target locations.…”