Video metamorphosis using dense flow fields

Metamorphosis is generally known as the continuous evolution of a source into a target. There is frame-to-frame camera or object motion most of the time when either the source or the target is an image sequence. It would be desirable to make smooth morphing transitions simultaneously along with the original motion. In this paper, we describe a novel semi-automatic morphing technique for image sequences. The proposed technique effectively exploits temporal coherence and automatic image matching. One-to-one dense mappings between pairs of corresponding frames are obtained by applying a compositing procedure and a hierarchical image matching technique based on dynamic programming. These dense mappings can be initialized by sparse frameto-frame feature correspondences obtained semi-automatically by integrating a friendly user interface with a robust feature tracking algorithm. Experimental results show that our approach to video metamorphosis can produce superior results.