Abstract
We propose Identical-pair Adversarial Networks (iPANs) to solve image-to-image translation problems, such as aerial-to-map, edge-to-photo, de-raining, and night-to-daytime. Our iPANs rely mainly on the effectiveness of adversarial loss function and its network architectures. Our iPANs consist of two main networks, an image transformation network T and a discriminative network D. We use U-NET for the transformation network T and a perceptual similarity network, which has two streams of VGG16 that share the same weights for network D. Our proposed adversarial losses play a minimax game against each other based on a real identical-pair and a fake identical-pair distinguished by the discriminative network D; e.g. a discriminative network D considers two inputs as a real pair only when they are identical, otherwise a fake pair. Meanwhile, the transformation network T tries to persuade the discriminator network D that the fake pair is a real pair. We experimented on several problems of image-to-image translation and achieved results that are comparable to those of some existing approaches, such as pix2pix, and PAN.
Highlights
Image-to-image translation refers to a constrained synthesis task of automatically transforming an input image to a synthetic image or mapping an input image to the desired output image
We evaluate the performance of our proposed Identical-pair Adversarial Networks (iPANs) on several image-to-image translation tasks, which are popular in the fields of image processing, computer vision, and computer graphics
We compared our results with some existing methods on several tasks depending on the best results for each approach, such as ID-cGANs, pix2pix, bicycleGANs, Perceptual Adversarial Networks (PAN), Dual-GAN, and cycleGAN
Summary
Image-to-image translation refers to a constrained synthesis task of automatically transforming an input image to a synthetic image or mapping an input image to the desired output image. There are several applications of image-to-image translation in the fields of image processing, computer graphics, and computer vision, that transform an input image and generate high-resolution RGB images. Convolutional neural networks have been used for various image-to-image translation tasks [6,7,8,9]. The convolutional neural networks were trained to find the mapping from the input image to the transformed image by penalizing the discrepancy between the output and ground-truth images. Those approaches achieved their performance differently depending on cost function designs. Producing sharp and realistic images is an open challenge and requires better approaches
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
