Recomposed realities: animating still images via patch clustering and randomness
This addresses the challenge of creating motion in static images for creative or entertainment applications, but it appears incremental as it builds on existing patch-based and clustering techniques.
The paper tackles the problem of animating still images by reconstructing them using patch clustering and random sampling from curated datasets, resulting in a method that emphasizes reinterpretation over replication.
We present a patch-based image reconstruction and animation method that uses existing image data to bring still images to life through motion. Image patches from curated datasets are grouped using k-means clustering and a new target image is reconstructed by matching and randomly sampling from these clusters. This approach emphasizes reinterpretation over replication, allowing the source and target domains to differ conceptually while sharing local structures.