IMAGINE: Image Synthesis by Image-Guided Model Inversion
This addresses the challenge of data-efficient image synthesis for users needing intuitive control, though it is incremental as it builds on existing inversion and GAN techniques.
The paper tackles the problem of generating high-quality and diverse images from only a single training sample by introducing IMAGINE, an inversion-based method that leverages a pre-trained classifier for semantic guidance and adversarial training, achieving favorable performance against state-of-the-art methods across object, scene, and texture domains.
We introduce an inversion based method, denoted as IMAge-Guided model INvErsion (IMAGINE), to generate high-quality and diverse images from only a single training sample. We leverage the knowledge of image semantics from a pre-trained classifier to achieve plausible generations via matching multi-level feature representations in the classifier, associated with adversarial training with an external discriminator. IMAGINE enables the synthesis procedure to simultaneously 1) enforce semantic specificity constraints during the synthesis, 2) produce realistic images without generator training, and 3) give users intuitive control over the generation process. With extensive experimental results, we demonstrate qualitatively and quantitatively that IMAGINE performs favorably against state-of-the-art GAN-based and inversion-based methods, across three different image domains (i.e., objects, scenes, and textures).