2022/02/25 by Zhenxing Zhang, Zhang, Zhenxing, Lambert Schomaker +1 · 2 citations
Computer Science · #Artificial intelligence #Classifier (UML) #Computer Vision and Pattern Recognition (cs.CV) #Computer science #Computer vision #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Generator (circuit theory) #Handwritten Text Recognition Techniques #Image (mathematics) #Image Processing and 3D Reconstruction #Machine Learning (cs.LG) #Pattern recognition (psychology) #Set (abstract data type) #cs.CV #cs.LG
paper · pdf · doi:10.48550/arxiv.2202.12929
published in arXiv (Cornell University) (Cornell University) · 18 pages
arxiv created 2022/02/25 · openalex publication_date 2022/02/25 · arxiv updated 2022/03/01 · openalex created_date 2022/04/03 · openalex updated_date 2026/07/28
Text-to-image generation intends to automatically produce a photo-realistic image, conditioned on a textual description. It can be potentially employed in the field of art creation, data augmentation, photo-editing, etc. Although many efforts have been dedicated to this task, it remains particularly challenging to generate believable, natural scenes. To facilitate the real-world applications of text-to-image synthesis, we focus on studying the following three issues: 1) How to ensure that generated samples are believable, realistic or natural? 2) How to exploit the latent space of the generator to edit a synthesized image? 3) How to improve the explainability of a text-to-image generation framework? In this work, we constructed two novel data sets (i.e., the Good & Bad bird and face data sets) consisting of successful as well as unsuccessful generated samples, according to strict criteria. To effectively and efficiently acquire high-quality images by increasing the probability of generating Good latent codes, we use a dedicated Good/Bad classifier for generated images. It is based on a pre-trained front end and fine-tuned on the basis of the proposed Good & Bad data set. After that, we present a novel algorithm which identifies semantically-understandable directions in the latent space of a conditional text-to-image GAN architecture by performing independent component analysis on the pre-trained weight values of the generator. Furthermore, we develop a background-flattening loss (BFL), to improve the background appearance in the edited image. Subsequently, we introduce linear interpolation analysis between pairs of keywords. This is extended into a similar triangular `linguistic' interpolation in order to take a deep look into what a text-to-image synthesis model has learned within the linguistic embeddings. Our data set is available at https://zenodo.org/record/6283798#.YhkNujMI2w.