Peizhuo Li (ETH Zurich) - Generative Motion Matching

The generation of natural, varied, and detailed motions is a core problem in computer animation. We present a generative model – generative motion matching – that “mines” as many diverse motions as possible from a single or few example sequences. In stark contrast to existing data-driven methods, which typically require long offline training time, are prone to visual artifacts, and tend to fail on large and complex skeletons, our method inherits the training-free nature and the superior quality of the well-known Motion Matching method. At the heart of our generative framework lies the generative motion matching module, which utilizes the bidirectional similarity as a generative cost function to motion matching, and operates in a multi-stage framework to progressively refine a random guess using exemplar motion matches. In addition to diverse motion generation, we show the versatility of our generative framework by extending it to a number of scenarios that are not possible with motion matching alone, including motion completion, key frame-guided generation, infinite looping, and motion reassembly.

Host: Rana Hanocka