Abstract
Realistic personalized avatars can play an important role in social interactions in virtual reality, increasing body ownership, presence, and dominance. A simple way to obtain the texture of an avatar is to use a single front-view image of a human and to generate the hidden back-view image. The realism of the generated image is crucial in improving the overall texture quality, and subjective image quality assessment methods can play an important role in the evaluation. The subjective methods, however, require dozens of human assessors, a controlled environment, and time. This paper proposes a deep learning-based image reality assessment method, which is fully automatic and has a short testing time of nearly a quarter second per image. We train various discriminators to predict whether an image is real or generated. The trained discriminators are then used to give a mean opinion score for the reality of an image. Through experiments on human back-view images, we show that our learning-based mean opinion scores are close to their subjective counterparts in terms of the root mean square error between them.
Highlights
Realistic personalized avatars increase body ownership, presence, and dominance in virtual environments [1,2], so they can play an important role in social interactions as well as applications like virtual dressing rooms
To quantify the similarity of the proposed method to the subjective mean opinion score (MOS) test, we performed a subjective test on selected back-view images
We have proposed learning-based MOS, which can be used to assess image reality
Summary
Realistic personalized avatars increase body ownership, presence, and dominance in virtual environments [1,2], so they can play an important role in social interactions as well as applications like virtual dressing rooms. Methods based on convolutional neural networks (CNNs) have been proposed to create a personalized avatar from single front-view images of humans [3,4], reducing the cost and space as well as simplifying the capturing process. We propose a fully automatic, cost-effective method for assessing the reality of a human back-view image. Shemelkov et al [12] proposed a method for evaluating the performance of class-conditional GANs. Shemelkov et al [12] proposed a method for evaluating the performance of class-conditional GANs In their method, by training an object-classification network using GAN-generated images and testing the classifier using real test images, ‘GAN-train’ accuracy is computed. By training the network using real images and testing the classifier using the generated images, ’GAN-test’ accuracy is computed.
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.
