Latent Image Animator: Learning to Animate Images via Latent Space Navigation
This work addresses a key limitation in image animation for applications like video generation, though it is incremental as it builds on existing generative models.
The paper tackles the problem of animating images when source images and driving videos have large appearance variations, and introduces the Latent Image Animator (LIA) that uses linear navigation in latent space to achieve this, outperforming state-of-the-art methods on datasets like VoxCeleb, Taichi, and TED-talk.
Due to the remarkable progress of deep generative models, animating images has become increasingly efficient, whereas associated results have become increasingly realistic. Current animation-approaches commonly exploit structure representation extracted from driving videos. Such structure representation is instrumental in transferring motion from driving videos to still images. However, such approaches fail in case the source image and driving video encompass large appearance variation. Moreover, the extraction of structure information requires additional modules that endow the animation-model with increased complexity. Deviating from such models, we here introduce the Latent Image Animator (LIA), a self-supervised autoencoder that evades need for structure representation. LIA is streamlined to animate images by linear navigation in the latent space. Specifically, motion in generated video is constructed by linear displacement of codes in the latent space. Towards this, we learn a set of orthogonal motion directions simultaneously, and use their linear combination, in order to represent any displacement in the latent space. Extensive quantitative and qualitative analysis suggests that our model systematically and significantly outperforms state-of-art methods on VoxCeleb, Taichi and TED-talk datasets w.r.t. generated quality.