Abstract
In this paper, we tackle the well-known problem of dataset construction from the point of its generation using generative adversarial networks (GAN). As semantic information of the dataset should have a proper alignment with images, controlling the image generation process of GAN comes to the first position. Considering this, we focus on conditioning the generative process by solely utilizing conditional information to achieve reliable control over the image generation. Unlike the existing works that consider the input (noise or image) in conjunction with conditions, our work considers transforming the input directly to the conditional space by utilizing the given conditions only. By doing so, we reveal the relations between conditions to determine their distinct and reliable feature space without the impact of input information. To fully leverage the conditional information, we propose a novel architectural framework (i.e., conditional transformation) that aims to learn features only from a set of conditions for guiding a generative model by transforming the input to the generator. Such an approach enables controlling the generator by setting its inputs according to the specific conditions necessary for semantically correct image generation. Given that the framework operates at the initial stage of generation, it can be plugged into any existing generative models and trained in an end-to-end manner together with the generator. Extensive experiments on various tasks, such as novel image synthesis and image-to-image translation, demonstrate that the conditional transformation of inputs facilitates solid control over the image generation process and thus shows its applicability for use in dataset construction.
Highlights
T HE dataset is a key element in teaching a learning-based method to understand real-world scenarios
EXPERIMENTS we present the performance of the proposed conditional transformation (CT) framework when used along with diverse existing generative adversarial networks (GAN) architectures for image synthesis and image-to-image translation tasks
The technical contribution of our work consists of controlling the image generation process by conditionally transforming the input to the generator such that it corresponds to the given conditions
Summary
T HE dataset is a key element in teaching a learning-based method to understand real-world scenarios. Several other methods [3], [17], [26], [34]– [37] introduce conditional information (occasionally with noise) into the hidden layers of the generator through normalization technique [38] by replacing the non-adaptive parameters (i.e., scale and shift) with input-dependent ones These parameters are learned based on conditions by utilizing the embedding functions. An example is the generation of facial images with multiple attributes corresponding to various genders, ages, and expression classes In consideration of these statements, we consider conditioning the generator from a different perspective where conditional information can be utilized on its own.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: 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.