Multi-Clip Video Editing from a Single Viewpoint Vineet Gandhi

We propose a framework for automatically generating multiple clips suitable for video editing by simulating pan-tilt-zoom camera move-ments within the frame of a single static camera. Assuming impor-tant actors and objects can be localized using computer vision tech-niques, our method requires only minimal user input to define the subject matter of each sub-clip. The composition of each sub-clip is automatically computed in a novel L1-norm optimization frame-work. Our approach encodes several common cinematographic practices into a single convex cost function minimization prob-lem, resulting in aesthetically pleasing sub-clips which can easily be edited together using off-the-shelf multi-clip video editing soft-ware. We demonstrate our approach on five video sequences of a live theatre performance by generating multiple synchronized sub-clips for each sequence