Point-based interactive editing serves as an essential tool to complement the controllability
of existing generative models. A concurrent work DragDiffusion updates the diffusion latent …

Image generation relies on massive training data that can hardly produce diverse images of
an unseen category according to a few examples. In this paper, we address this dilemma by …

The generalization with respect to domain shifts, as they frequently appear in applications
such as autonomous driving, is one of the remaining big challenges for deep learning …

GAN inversion is indispensable for applying the powerful editability of GAN to real images.
However, existing methods invert video frames individually often leading to undesired …

Throughout history, static paintings have captivated viewers within display frames, yet the
possibility of making these masterpieces vividly interactive remains intriguing. This research …

Abstract Generative Neural Radiance Fields (NeRFs) have demonstrated remarkable
proficiency in synthesizing multi-view images by learning the distribution of a set of unposed …

StyleGAN's disentangled style representation enables powerful image editing by
manipulating the latent variables, but accurately mapping real-world images to their latent …

Training diffusion models on limited datasets poses challenges in terms of limited
generation capacity and expressiveness, leading to unsatisfactory results in various …

Abstract The fidelity of Generative Adversarial Networks (GAN) inversion is impeded by Out-
Of-Domain (OOD) areas (eg, background, accessories) in the image. Detecting the OOD …

In this paper, we introduce\(\textrm {D}^ 4\)-VTON, an innovative solution for image-based
virtual try-on. We address challenges from previous studies, such as semantic …